{"licence": "Creative Commons Attribution 4.0 International (CC BY 4.0)", "licence_url": "https://creativecommons.org/licenses/by/4.0/", "attribution": "Spring Prompt (springprompt.com)", "source_url": "https://springprompt.com/benchmarks/deckbench", "release": "2026-11-10-cad1a1ab581e", "test_your_prompt": "These are our tasks, not yours. Test your own prompt on these models: https://springprompt.com/prompt-test", "results": [{"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 1592.5, "ci_low": 1482.5, "ci_high": 1765.8, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 1475.0, "ci_low": 1412.6, "ci_high": 1583.8, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 1424.9, "ci_low": 1318.5, "ci_high": 1571.6, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 1420.1, "ci_low": 1348.2, "ci_high": 1522.8, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 1376.8, "ci_low": 1300.1, "ci_high": 1469.2, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 1359.0, "ci_low": 1275.5, "ci_high": 1455.3, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 1331.7, "ci_low": 1229.5, "ci_high": 1443.7, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 1291.2, "ci_low": 1235.9, "ci_high": 1368.5, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 1288.7, "ci_low": 1188.2, "ci_high": 1390.9, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 1284.9, "ci_low": 1227.0, "ci_high": 1368.8, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 1279.7, "ci_low": 1208.8, "ci_high": 1363.2, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 1239.9, "ci_low": 1123.7, "ci_high": 1371.4, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 1237.5, "ci_low": 1170.6, "ci_high": 1302.5, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 1222.0, "ci_low": 1147.3, "ci_high": 1307.1, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 1155.3, "ci_low": 1084.3, "ci_high": 1233.5, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 1149.6, "ci_low": 1017.3, "ci_high": 1256.1, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 1100.7, "ci_low": 1000.2, "ci_high": 1208.9, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 1079.7, "ci_low": 973.2, "ci_high": 1163.3, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 1060.2, "ci_low": 999.2, "ci_high": 1140.6, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 984.5, "ci_low": 912.6, "ci_high": 1058.3, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 842.6, "ci_low": 775.1, "ci_high": 888.2, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 822.6, "ci_low": 680.5, "ci_high": 936.6, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 813.4, "ci_low": 746.9, "ci_high": 871.1, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 760.0, "ci_low": 665.7, "ci_high": 851.8, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 742.1, "ci_low": 663.4, "ci_high": 802.1, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 681.0, "ci_low": 595.4, "ci_high": 756.7, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 640.0, "ci_low": 542.1, "ci_high": 724.1, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 632.1, "ci_low": 520.8, "ci_high": 731.7, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 619.5, "ci_low": 525.6, "ci_high": 700.9, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "rating", "metric_name": "DeckBench: deck rating", "unit": "rating", "direction": "higher", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 564.8, "ci_low": 427.8, "ci_high": 635.2, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "presentable_rate", "metric_name": "DeckBench: presentable decks", "unit": "% of tasks", "direction": "higher", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "accurate_rate", "metric_name": "DeckBench: accurate decks", "unit": "% of tasks", "direction": "higher", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "layout_clean_rate", "metric_name": "DeckBench: clean layout", "unit": "% of tasks", "direction": "higher", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 88.51, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 87.9, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 83.89, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 81.38, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 80.75, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 78.47, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 78.1, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 77.83, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 76.44, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 71.96, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 71.39, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 68.47, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 67.11, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 66.95, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 66.91, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 66.62, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 61.71, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 61.06, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 60.22, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 58.49, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 56.05, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 55.2, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 54.37, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 51.39, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 47.62, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 46.23, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 43.65, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 40.87, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 30.51, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "design_quality", "metric_name": "DeckBench: design quality", "unit": "% of the maximum", "direction": "higher", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 29.3, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 89.74, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 88.14, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 86.57, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 84.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 82.14, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 77.97, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 77.51, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 75.95, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 71.08, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 70.8, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 70.35, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 63.04, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 62.57, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 60.84, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 60.83, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 51.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 50.55, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 48.62, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 48.39, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 48.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 45.99, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 42.98, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 40.46, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 39.56, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 32.28, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 30.19, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 28.66, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 25.76, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 25.23, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "win_rate", "metric_name": "DeckBench: head-to-head win rate", "unit": "% of comparisons", "direction": "higher", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 24.43, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 0.00857, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 0.00865, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 0.00905, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.01042, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 0.01162, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 0.01262, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 0.01477, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 0.02552, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 0.0359, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.0448, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 0.04607, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 0.04626, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 0.0465, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 0.06134, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 0.06722, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 0.11468, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 0.1355, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 0.14527, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 0.15273, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 0.15295, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.18211, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 0.20966, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 0.25243, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 0.26102, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 0.4537, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.54068, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 0.80661, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 1.09215, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "cost_per_deck", "metric_name": "DeckBench: cost per deck", "unit": "US dollars", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 1.39571, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 15.2, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 25.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 28.6, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 51.7, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 62.1, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 67.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 73.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 78.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 83.3, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 83.4, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 89.4, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 93.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 98.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 100.6, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 114.2, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 134.9, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 154.4, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 157.9, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 169.1, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 174.8, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 178.9, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 217.3, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 285.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 311.4, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 330.1, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 332.3, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 374.8, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 437.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 543.1, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "seconds_per_deck", "metric_name": "DeckBench: time per deck", "unit": "seconds", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 1288.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 12086.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 16952.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 21019.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 27796.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 32448.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 33285.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 33765.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 45141.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "tokens_per_deck", "metric_name": "DeckBench: output tokens per deck", "unit": "tokens", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 51739.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.layout", "metric_name": "DeckBench: layout defects", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.draft", "metric_name": "DeckBench: draft figure quoted", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.numbers", "metric_name": "DeckBench: unsupported numbers", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.findings", "metric_name": "DeckBench: findings missing", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.recommendation", "metric_name": "DeckBench: recommendation late or wrong", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.caveats", "metric_name": "DeckBench: caveat dropped", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.misleading", "metric_name": "DeckBench: misleading metric used", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.claims", "metric_name": "DeckBench: unsupported claims", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-5-flash-lite", "model_name": "Gemini 3.5 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.5 Flash-Lite", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "xai-grok-4-7", "model_name": "Grok 4.7", "developer": "xAI", "configuration": "Grok 4.7", "value": 16.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "particle-ai-particle-1", "model_name": "Particle-1", "developer": "Particle AI", "configuration": "Particle-1", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "alibaba-qwen3-8-max-0902", "model_name": "Qwen3.8-Max (0902)", "developer": "Alibaba", "configuration": "Qwen3.8-Max (0902)", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "zai-glm-5-3-flashx-legal", "model_name": "GLM-5.3 FlashX Legal", "developer": "Z.ai", "configuration": "GLM-5.3 FlashX Legal", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "deepseek-deepseek-v4-pro", "model_name": "DeepSeek-V4-Pro (0423)", "developer": "DeepSeek", "configuration": "DeepSeek-V4-Pro (0423)", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}, {"benchmark": "deckbench", "metric": "fail.design", "metric_name": "DeckBench: slides needing work", "unit": "% of tasks", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 100.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "6", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": "GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority"}]}