{"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/bulletbench", "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": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "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": 892.0, "ci_low": 641.0, "ci_high": 1081.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "google-gemini-3-1-flash-lite", "model_name": "Gemini 3.1 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.1 Flash-Lite (minimal reasoning)", "value": 780.0, "ci_low": 589.0, "ci_high": 940.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "google-gemini-3-5-flash", "model_name": "Gemini 3.5 Flash", "developer": "Google", "configuration": "Gemini 3.5 Flash (minimal reasoning)", "value": 727.0, "ci_low": 567.0, "ci_high": 890.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "inception-mercury-decide", "model_name": "Mercury Decide", "developer": "Inception", "configuration": "Mercury Decide", "value": 614.0, "ci_low": 470.0, "ci_high": 766.0, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast (low reasoning)", "value": 584.0, "ci_low": 327.0, "ci_high": 806.0, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "liquid-d1", "model_name": "d1", "developer": "Liquid AI", "configuration": "d1", "value": 572.0, "ci_low": 428.0, "ci_high": 718.0, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "bytedance-seed-2-0-mini", "model_name": "Seed-2.0-Mini", "developer": "ByteDance", "configuration": "Seed-2.0-Mini (minimal reasoning)", "value": 571.0, "ci_low": 420.0, "ci_high": 710.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "mistral-mistral-small-2603", "model_name": "Mistral Small 4", "developer": "Mistral AI", "configuration": "Mistral Small 4 (no reasoning)", "value": 571.0, "ci_low": 420.0, "ci_high": 699.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-5-4-mini", "model_name": "GPT-5.4 mini", "developer": "OpenAI", "configuration": "GPT-5.4 mini (no reasoning)", "value": 571.0, "ci_low": 412.0, "ci_high": 721.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "thinkingmachines-inkling-small", "model_name": "Inkling Small", "developer": "Thinking Machines", "configuration": "Inkling Small (no reasoning)", "value": 571.0, "ci_low": 428.0, "ci_high": 698.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "typesafe-jev-1-13", "model_name": "Jev 1.13", "developer": "TypeSafe", "configuration": "Jev 1.13", "value": 571.0, "ci_low": 429.0, "ci_high": 719.0, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3 (no reasoning)", "value": 555.0, "ci_low": 263.0, "ci_high": 727.0, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "cloudflare-clef-flash", "model_name": "Clef Flash", "developer": "Cloudflare", "configuration": "Clef Flash", "value": 529.0, "ci_low": 388.0, "ci_high": 670.0, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "cloudflare-clef", "model_name": "Clef", "developer": "Cloudflare", "configuration": "Clef", "value": 529.0, "ci_low": 373.0, "ci_high": 653.0, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5 (no reasoning)", "value": 529.0, "ci_low": 361.0, "ci_high": 669.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 529.0, "ci_low": 308.0, "ci_high": 680.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-luna-decisions", "model_name": "GPT-6 Luna Decisions", "developer": "OpenAI", "configuration": "GPT-6 Luna Decisions", "value": 529.0, "ci_low": 407.0, "ci_high": 643.0, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra (low reasoning)", "value": 491.0, "ci_low": 0.0, "ci_high": 727.0, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "perplexity-pplx-decider-v1-27b", "model_name": "Decider V1 27B", "developer": "Perplexity", "configuration": "Decider V1 27B", "value": 420.0, "ci_low": 367.0, "ci_high": 488.0, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "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 (low reasoning)", "value": 310.0, "ci_low": 0.0, "ci_high": 641.0, "observed_on": "2026-11-04", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 249.0, "ci_low": 0.0, "ci_high": 486.0, "observed_on": "2026-10-06", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Mistral API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash (low reasoning)", "value": 197.0, "ci_low": 0.0, "ci_high": 439.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-5-4-nano", "model_name": "GPT-5.4 nano", "developer": "OpenAI", "configuration": "GPT-5.4 nano (no reasoning)", "value": 197.0, "ci_low": 0.0, "ci_high": 425.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast", "value": 197.0, "ci_low": 0.0, "ci_high": 404.0, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 33.0, "ci_low": 0.0, "ci_high": 248.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5 (low reasoning)", "value": 33.0, "ci_low": 0.0, "ci_high": 270.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "google-gemini-3-6-flash", "model_name": "Gemini 3.6 Flash", "developer": "Google", "configuration": "Gemini 3.6 Flash (minimal reasoning)", "value": 33.0, "ci_low": 0.0, "ci_high": 248.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.ladder_elo", "metric_name": "BulletBench Lightning 10+1: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "jaredpalmer-kev-4b", "model_name": "Kev 4B", "developer": "Jared Palmer", "configuration": "Kev 4B", "value": 33.0, "ci_low": 0.0, "ci_high": 270.0, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "bytedance-seed-2-0-mini", "model_name": "Seed-2.0-Mini", "developer": "ByteDance", "configuration": "Seed-2.0-Mini (minimal reasoning)", "value": 272.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "inception-mercury-decide", "model_name": "Mercury Decide", "developer": "Inception", "configuration": "Mercury Decide", "value": 368.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3 (no reasoning)", "value": 421.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 512.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "typesafe-jev-1-13", "model_name": "Jev 1.13", "developer": "TypeSafe", "configuration": "Jev 1.13", "value": 515.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-luna-decisions", "model_name": "GPT-6 Luna Decisions", "developer": "OpenAI", "configuration": "GPT-6 Luna Decisions", "value": 517.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5 (no reasoning)", "value": 523.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "cloudflare-clef-flash", "model_name": "Clef Flash", "developer": "Cloudflare", "configuration": "Clef Flash", "value": 543.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "mistral-mistral-small-2603", "model_name": "Mistral Small 4", "developer": "Mistral AI", "configuration": "Mistral Small 4 (no reasoning)", "value": 569.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "thinkingmachines-inkling-small", "model_name": "Inkling Small", "developer": "Thinking Machines", "configuration": "Inkling Small (no reasoning)", "value": 586.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "perplexity-pplx-decider-v1-27b", "model_name": "Decider V1 27B", "developer": "Perplexity", "configuration": "Decider V1 27B", "value": 594.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "liquid-d1", "model_name": "d1", "developer": "Liquid AI", "configuration": "d1", "value": 615.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "cloudflare-clef", "model_name": "Clef", "developer": "Cloudflare", "configuration": "Clef", "value": 713.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "google-gemini-3-1-flash-lite", "model_name": "Gemini 3.1 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.1 Flash-Lite (minimal reasoning)", "value": 767.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 826.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Mistral API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "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": 882.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "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 (low reasoning)", "value": 886.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-5-4-mini", "model_name": "GPT-5.4 mini", "developer": "OpenAI", "configuration": "GPT-5.4 mini (no reasoning)", "value": 943.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "jaredpalmer-kev-4b", "model_name": "Kev 4B", "developer": "Jared Palmer", "configuration": "Kev 4B", "value": 1024.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-5-4-nano", "model_name": "GPT-5.4 nano", "developer": "OpenAI", "configuration": "GPT-5.4 nano (no reasoning)", "value": 1048.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "google-gemini-3-5-flash", "model_name": "Gemini 3.5 Flash", "developer": "Google", "configuration": "Gemini 3.5 Flash (minimal reasoning)", "value": 1178.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast (low reasoning)", "value": 1241.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra (low reasoning)", "value": 1261.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 1279.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast", "value": 1370.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "google-gemini-3-6-flash", "model_name": "Gemini 3.6 Flash", "developer": "Google", "configuration": "Gemini 3.6 Flash (minimal reasoning)", "value": 1520.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash (low reasoning)", "value": 1844.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.median_latency_ms", "metric_name": "BulletBench Lightning 10+1: median move time", "unit": "milliseconds", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5 (low reasoning)", "value": 2083.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "bytedance-seed-2-0-mini", "model_name": "Seed-2.0-Mini", "developer": "ByteDance", "configuration": "Seed-2.0-Mini (minimal reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "cloudflare-clef-flash", "model_name": "Clef Flash", "developer": "Cloudflare", "configuration": "Clef Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "cloudflare-clef", "model_name": "Clef", "developer": "Cloudflare", "configuration": "Clef", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "google-gemini-3-1-flash-lite", "model_name": "Gemini 3.1 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.1 Flash-Lite (minimal reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "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": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3 (no reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "inception-mercury-decide", "model_name": "Mercury Decide", "developer": "Inception", "configuration": "Mercury Decide", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "liquid-d1", "model_name": "d1", "developer": "Liquid AI", "configuration": "d1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5 (no reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "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": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "mistral-mistral-small-2603", "model_name": "Mistral Small 4", "developer": "Mistral AI", "configuration": "Mistral Small 4 (no reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-5-4-mini", "model_name": "GPT-5.4 mini", "developer": "OpenAI", "configuration": "GPT-5.4 mini (no reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-luna-decisions", "model_name": "GPT-6 Luna Decisions", "developer": "OpenAI", "configuration": "GPT-6 Luna Decisions", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "perplexity-pplx-decider-v1-27b", "model_name": "Decider V1 27B", "developer": "Perplexity", "configuration": "Decider V1 27B", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "thinkingmachines-inkling-small", "model_name": "Inkling Small", "developer": "Thinking Machines", "configuration": "Inkling Small (no reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "typesafe-jev-1-13", "model_name": "Jev 1.13", "developer": "TypeSafe", "configuration": "Jev 1.13", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "google-gemini-3-5-flash", "model_name": "Gemini 3.5 Flash", "developer": "Google", "configuration": "Gemini 3.5 Flash (minimal reasoning)", "value": 33.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast (low reasoning)", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 58.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Mistral API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra (low reasoning)", "value": 62.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "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 (low reasoning)", "value": 62.5, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "jaredpalmer-kev-4b", "model_name": "Kev 4B", "developer": "Jared Palmer", "configuration": "Kev 4B", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-5-4-nano", "model_name": "GPT-5.4 nano", "developer": "OpenAI", "configuration": "GPT-5.4 nano (no reasoning)", "value": 75.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash (low reasoning)", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5 (low reasoning)", "value": 91.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.time_loss_rate", "metric_name": "BulletBench Lightning 10+1: games lost on time", "unit": "% of games", "direction": "lower", "model": "google-gemini-3-6-flash", "model_name": "Gemini 3.6 Flash", "developer": "Google", "configuration": "Gemini 3.6 Flash (minimal reasoning)", "value": 91.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "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": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5 (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "cloudflare-clef-flash", "model_name": "Clef Flash", "developer": "Cloudflare", "configuration": "Clef Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "cloudflare-clef", "model_name": "Clef", "developer": "Cloudflare", "configuration": "Clef", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "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": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "google-gemini-3-6-flash", "model_name": "Gemini 3.6 Flash", "developer": "Google", "configuration": "Gemini 3.6 Flash (minimal reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "inception-mercury-decide", "model_name": "Mercury Decide", "developer": "Inception", "configuration": "Mercury Decide", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "jaredpalmer-kev-4b", "model_name": "Kev 4B", "developer": "Jared Palmer", "configuration": "Kev 4B", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "liquid-d1", "model_name": "d1", "developer": "Liquid AI", "configuration": "d1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5 (no reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-5-4-mini", "model_name": "GPT-5.4 mini", "developer": "OpenAI", "configuration": "GPT-5.4 mini (no reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-6-luna-decisions", "model_name": "GPT-6 Luna Decisions", "developer": "OpenAI", "configuration": "GPT-6 Luna Decisions", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "perplexity-pplx-decider-v1-27b", "model_name": "Decider V1 27B", "developer": "Perplexity", "configuration": "Decider V1 27B", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "typesafe-jev-1-13", "model_name": "Jev 1.13", "developer": "TypeSafe", "configuration": "Jev 1.13", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "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 (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "google-gemini-3-1-flash-lite", "model_name": "Gemini 3.1 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.1 Flash-Lite (minimal reasoning)", "value": 0.2, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.27, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3 (no reasoning)", "value": 0.35, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "google-gemini-3-5-flash", "model_name": "Gemini 3.5 Flash", "developer": "Google", "configuration": "Gemini 3.5 Flash (minimal reasoning)", "value": 0.6, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "bytedance-seed-2-0-mini", "model_name": "Seed-2.0-Mini", "developer": "ByteDance", "configuration": "Seed-2.0-Mini (minimal reasoning)", "value": 0.8, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "mistral-mistral-small-2603", "model_name": "Mistral Small 4", "developer": "Mistral AI", "configuration": "Mistral Small 4 (no reasoning)", "value": 0.99, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 1.84, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Mistral API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-5-4-nano", "model_name": "GPT-5.4 nano", "developer": "OpenAI", "configuration": "GPT-5.4 nano (no reasoning)", "value": 4.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.invalid_move_rate", "metric_name": "BulletBench Lightning 10+1: invalid moves", "unit": "% of moves", "direction": "lower", "model": "thinkingmachines-inkling-small", "model_name": "Inkling Small", "developer": "Thinking Machines", "configuration": "Inkling Small (no reasoning)", "value": 5.26, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "inception-mercury-decide", "model_name": "Mercury Decide", "developer": "Inception", "configuration": "Mercury Decide", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "liquid-d1", "model_name": "d1", "developer": "Liquid AI", "configuration": "d1", "value": 0.00045, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "jaredpalmer-kev-4b", "model_name": "Kev 4B", "developer": "Jared Palmer", "configuration": "Kev 4B", "value": 0.000635, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "perplexity-pplx-decider-v1-27b", "model_name": "Decider V1 27B", "developer": "Perplexity", "configuration": "Decider V1 27B", "value": 0.000869, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3 (no reasoning)", "value": 0.001035, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "typesafe-jev-1-13", "model_name": "Jev 1.13", "developer": "TypeSafe", "configuration": "Jev 1.13", "value": 0.001382, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "bytedance-seed-2-0-mini", "model_name": "Seed-2.0-Mini", "developer": "ByteDance", "configuration": "Seed-2.0-Mini (minimal reasoning)", "value": 0.001925, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "cloudflare-clef-flash", "model_name": "Clef Flash", "developer": "Cloudflare", "configuration": "Clef Flash", "value": 0.002391, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-5-4-nano", "model_name": "GPT-5.4 nano", "developer": "OpenAI", "configuration": "GPT-5.4 nano (no reasoning)", "value": 0.002494, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-luna-decisions", "model_name": "GPT-6 Luna Decisions", "developer": "OpenAI", "configuration": "GPT-6 Luna Decisions", "value": 0.002502, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "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 (low reasoning)", "value": 0.003521, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash (low reasoning)", "value": 0.004053, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "mistral-mistral-small-2603", "model_name": "Mistral Small 4", "developer": "Mistral AI", "configuration": "Mistral Small 4 (no reasoning)", "value": 0.004091, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "google-gemini-3-6-flash", "model_name": "Gemini 3.6 Flash", "developer": "Google", "configuration": "Gemini 3.6 Flash (minimal reasoning)", "value": 0.004995, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "google-gemini-3-1-flash-lite", "model_name": "Gemini 3.1 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.1 Flash-Lite (minimal reasoning)", "value": 0.005835, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra (low reasoning)", "value": 0.006417, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "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": 0.006979, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "cloudflare-clef", "model_name": "Clef", "developer": "Cloudflare", "configuration": "Clef", "value": 0.007186, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "thinkingmachines-inkling-small", "model_name": "Inkling Small", "developer": "Thinking Machines", "configuration": "Inkling Small (no reasoning)", "value": 0.007678, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5 (low reasoning)", "value": 0.008041, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "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": 0.009534, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 0.011659, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Mistral API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-5-4-mini", "model_name": "GPT-5.4 mini", "developer": "OpenAI", "configuration": "GPT-5.4 mini (no reasoning)", "value": 0.013619, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "google-gemini-3-5-flash", "model_name": "Gemini 3.5 Flash", "developer": "Google", "configuration": "Gemini 3.5 Flash (minimal reasoning)", "value": 0.021301, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "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": 0.029713, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "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 (no reasoning)", "value": 0.049138, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast", "value": 0.158258, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "10p1.cost_per_game", "metric_name": "BulletBench Lightning 10+1: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast (low reasoning)", "value": 0.165374, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 1147.0, "ci_low": 979.0, "ci_high": 1373.0, "observed_on": "2026-10-13", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 1126.0, "ci_low": 949.0, "ci_high": 1347.0, "observed_on": "2026-09-30", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 1100.0, "ci_low": 789.0, "ci_high": 1450.0, "observed_on": "2026-11-10", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast", "value": 927.0, "ci_low": 657.0, "ci_high": 1152.0, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 839.0, "ci_low": 595.0, "ci_high": 1072.0, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "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": 707.0, "ci_low": 458.0, "ci_high": 933.0, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "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": 683.0, "ci_low": 487.0, "ci_high": 1130.0, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 683.0, "ci_low": 487.0, "ci_high": 1130.0, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 645.0, "ci_low": 0.0, "ci_high": 890.0, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 557.0, "ci_low": 370.0, "ci_high": 727.0, "observed_on": "2026-10-08", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 557.0, "ci_low": 370.0, "ci_high": 727.0, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 557.0, "ci_low": 370.0, "ci_high": 727.0, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 557.0, "ci_low": 370.0, "ci_high": 727.0, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 555.0, "ci_low": 392.0, "ci_high": 727.0, "observed_on": "2026-10-06", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Mistral API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 555.0, "ci_low": 396.0, "ci_high": 765.0, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 491.0, "ci_low": 299.0, "ci_high": 659.0, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 358.0, "ci_low": 0.0, "ci_high": 619.0, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.ladder_elo", "metric_name": "BulletBench Blitz 3+2: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 358.0, "ci_low": 0.0, "ci_high": 600.0, "observed_on": "2026-10-12", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 519.5, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "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": 789.0, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 837.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Mistral API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 1297.0, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 1937.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast", "value": 2112.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 3722.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 4039.0, "ci_low": null, "ci_high": null, "observed_on": "2026-09-30", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 4783.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 4858.5, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 5365.0, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 5612.0, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 6172.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "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": 6219.0, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 6979.0, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 7106.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 7513.0, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.median_latency_ms", "metric_name": "BulletBench Blitz 3+2: median move time", "unit": "milliseconds", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 8122.0, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "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": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "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": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "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": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "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": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Mistral API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "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": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 12.5, "ci_low": null, "ci_high": null, "observed_on": "2026-09-30", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra", "value": 25.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "direction": "lower", "model": "anthropic-claude-fable-5-1", "model_name": "Claude Fable 5.1", "developer": "Anthropic", "configuration": "Claude Fable 5.1", "value": 37.5, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 37.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "direction": "lower", "model": "anthropic-claude-sonnet-5-5", "model_name": "Claude Sonnet 5.5", "developer": "Anthropic", "configuration": "Claude Sonnet 5.5", "value": 37.5, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 37.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 37.5, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "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": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "direction": "lower", "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": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "direction": "lower", "model": "anthropic-claude-opus-5-5", "model_name": "Claude Opus 5.5", "developer": "Anthropic", "configuration": "Claude Opus 5.5", "value": 62.5, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.time_loss_rate", "metric_name": "BulletBench Blitz 3+2: games lost on time", "unit": "% of games", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 75.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "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": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "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": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "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": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "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": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "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": "2026-09-30", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "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": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "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": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "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": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "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": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "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": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "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": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.41, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "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": 0.53, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 0.56, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Mistral API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.invalid_move_rate", "metric_name": "BulletBench Blitz 3+2: invalid moves", "unit": "% of moves", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 0.86, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna", "value": 0.007164, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "unit": "US dollars", "direction": "lower", "model": "mistral-mistral-large-4-0", "model_name": "Mistral Large 4", "developer": "Mistral AI", "configuration": "Mistral Large 4", "value": 0.008238, "ci_low": null, "ci_high": null, "observed_on": "2026-10-06", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Mistral API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "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": 0.008653, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "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": 0.01643, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "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": 0.026106, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "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": 0.031778, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "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": 0.070617, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "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": 0.076745, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "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": 0.079469, "ci_low": null, "ci_high": null, "observed_on": "2026-09-30", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "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": 0.115847, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "unit": "US dollars", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 0.12185, "ci_low": null, "ci_high": null, "observed_on": "2026-10-13", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "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": 0.175778, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "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": 0.207138, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "unit": "US dollars", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder", "value": 0.267174, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.284697, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "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": 0.384659, "ci_low": null, "ci_high": null, "observed_on": "2026-09-29", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "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": 0.523922, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "180p2.cost_per_game", "metric_name": "BulletBench Blitz 3+2: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast", "value": 1.098534, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash (low reasoning)", "value": 1068.0, "ci_low": 888.0, "ci_high": 1241.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast (low reasoning)", "value": 928.0, "ci_low": 676.0, "ci_high": 1184.0, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "google-gemini-3-1-flash-lite", "model_name": "Gemini 3.1 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.1 Flash-Lite (minimal reasoning)", "value": 912.0, "ci_low": 697.0, "ci_high": 1087.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "google-gemini-3-5-flash", "model_name": "Gemini 3.5 Flash", "developer": "Google", "configuration": "Gemini 3.5 Flash (minimal reasoning)", "value": 810.0, "ci_low": 582.0, "ci_high": 1006.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "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": 780.0, "ci_low": 586.0, "ci_high": 956.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "google-gemini-3-6-flash", "model_name": "Gemini 3.6 Flash", "developer": "Google", "configuration": "Gemini 3.6 Flash (minimal reasoning)", "value": 628.0, "ci_low": 429.0, "ci_high": 849.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "liquid-d1", "model_name": "d1", "developer": "Liquid AI", "configuration": "d1", "value": 584.0, "ci_low": 435.0, "ci_high": 749.0, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "inception-mercury-decide", "model_name": "Mercury Decide", "developer": "Inception", "configuration": "Mercury Decide", "value": 572.0, "ci_low": 428.0, "ci_high": 717.0, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5 (no reasoning)", "value": 572.0, "ci_low": 428.0, "ci_high": 708.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "typesafe-jev-1-13", "model_name": "Jev 1.13", "developer": "TypeSafe", "configuration": "Jev 1.13", "value": 572.0, "ci_low": 429.0, "ci_high": 720.0, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast", "value": 540.0, "ci_low": 296.0, "ci_high": 765.0, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "cloudflare-clef-flash", "model_name": "Clef Flash", "developer": "Cloudflare", "configuration": "Clef Flash", "value": 529.0, "ci_low": 388.0, "ci_high": 670.0, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "cloudflare-clef", "model_name": "Clef", "developer": "Cloudflare", "configuration": "Clef", "value": 529.0, "ci_low": 373.0, "ci_high": 653.0, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview (low reasoning)", "value": 529.0, "ci_low": 329.0, "ci_high": 698.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 529.0, "ci_low": 263.0, "ci_high": 688.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "thinkingmachines-inkling-small", "model_name": "Inkling Small", "developer": "Thinking Machines", "configuration": "Inkling Small (no reasoning)", "value": 529.0, "ci_low": 393.0, "ci_high": 652.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2 (low reasoning)", "value": 491.0, "ci_low": 230.0, "ci_high": 714.0, "observed_on": "2026-11-10", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 491.0, "ci_low": 187.0, "ci_high": 738.0, "observed_on": "2026-11-10", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3 (no reasoning)", "value": 491.0, "ci_low": 299.0, "ci_high": 659.0, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-luna-decisions", "model_name": "GPT-6 Luna Decisions", "developer": "OpenAI", "configuration": "GPT-6 Luna Decisions", "value": 489.0, "ci_low": 376.0, "ci_high": 608.0, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "bytedance-seed-2-0-mini", "model_name": "Seed-2.0-Mini", "developer": "ByteDance", "configuration": "Seed-2.0-Mini (minimal reasoning)", "value": 453.0, "ci_low": 355.0, "ci_high": 538.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 445.0, "ci_low": 135.0, "ci_high": 626.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol (no reasoning)", "value": 445.0, "ci_low": 215.0, "ci_high": 639.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder (minimal reasoning)", "value": 409.0, "ci_low": 100.0, "ci_high": 609.0, "observed_on": "2026-10-12", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 385.0, "ci_low": 276.0, "ci_high": 464.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "perplexity-pplx-decider-v1-27b", "model_name": "Decider V1 27B", "developer": "Perplexity", "configuration": "Decider V1 27B", "value": 385.0, "ci_low": 279.0, "ci_high": 459.0, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "mistral-mistral-small-2603", "model_name": "Mistral Small 4", "developer": "Mistral AI", "configuration": "Mistral Small 4 (no reasoning)", "value": 367.0, "ci_low": 126.0, "ci_high": 564.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra (low reasoning)", "value": 358.0, "ci_low": 0.0, "ci_high": 595.0, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "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 (low reasoning)", "value": 358.0, "ci_low": 0.0, "ci_high": 600.0, "observed_on": "2026-11-04", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol (low reasoning)", "value": 323.0, "ci_low": 51.0, "ci_high": 529.0, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra (low reasoning)", "value": 323.0, "ci_low": 0.0, "ci_high": 528.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5 (no reasoning)", "value": 310.0, "ci_low": 0.0, "ci_high": 575.0, "observed_on": "2026-10-08", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna (no reasoning)", "value": 249.0, "ci_low": 0.0, "ci_high": 453.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-5-4-mini", "model_name": "GPT-5.4 mini", "developer": "OpenAI", "configuration": "GPT-5.4 mini (no reasoning)", "value": 197.0, "ci_low": 0.0, "ci_high": 445.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 197.0, "ci_low": 0.0, "ci_high": 439.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017 (low reasoning)", "value": 147.0, "ci_low": 0.0, "ci_high": 443.0, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 147.0, "ci_low": 0.0, "ci_high": 381.0, "observed_on": "2026-10-14", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3 (low reasoning)", "value": 147.0, "ci_low": 0.0, "ci_high": 381.0, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro (low reasoning)", "value": 147.0, "ci_low": 0.0, "ci_high": 381.0, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash (low reasoning)", "value": 115.0, "ci_low": 0.0, "ci_high": 485.0, "observed_on": "2026-11-04", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3 (minimal reasoning)", "value": 33.0, "ci_low": 0.0, "ci_high": 270.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3 (low reasoning)", "value": 33.0, "ci_low": 0.0, "ci_high": 270.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-5-4-nano", "model_name": "GPT-5.4 nano", "developer": "OpenAI", "configuration": "GPT-5.4 nano (no reasoning)", "value": 33.0, "ci_low": 0.0, "ci_high": 273.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 33.0, "ci_low": 0.0, "ci_high": 270.0, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "zai-glm-5-3", "model_name": "GLM-5.3", "developer": "Z.ai", "configuration": "GLM-5.3 (low reasoning)", "value": 33.0, "ci_low": 0.0, "ci_high": 270.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.ladder_elo", "metric_name": "BulletBench Bullet 60s: ladder Elo", "unit": "ladder Elo", "direction": "higher", "model": "zai-glm-5-3", "model_name": "GLM-5.3", "developer": "Z.ai", "configuration": "GLM-5.3", "value": 33.0, "ci_low": 0.0, "ci_high": 270.0, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "bytedance-seed-2-0-mini", "model_name": "Seed-2.0-Mini", "developer": "ByteDance", "configuration": "Seed-2.0-Mini (minimal reasoning)", "value": 276.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "inception-mercury-decide", "model_name": "Mercury Decide", "developer": "Inception", "configuration": "Mercury Decide", "value": 367.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3 (no reasoning)", "value": 423.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "typesafe-jev-1-13", "model_name": "Jev 1.13", "developer": "TypeSafe", "configuration": "Jev 1.13", "value": 508.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-luna-decisions", "model_name": "GPT-6 Luna Decisions", "developer": "OpenAI", "configuration": "GPT-6 Luna Decisions", "value": 518.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5 (no reasoning)", "value": 519.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "cloudflare-clef-flash", "model_name": "Clef Flash", "developer": "Cloudflare", "configuration": "Clef Flash", "value": 522.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 524.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "thinkingmachines-inkling-small", "model_name": "Inkling Small", "developer": "Thinking Machines", "configuration": "Inkling Small (no reasoning)", "value": 565.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "mistral-mistral-small-2603", "model_name": "Mistral Small 4", "developer": "Mistral AI", "configuration": "Mistral Small 4 (no reasoning)", "value": 570.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "perplexity-pplx-decider-v1-27b", "model_name": "Decider V1 27B", "developer": "Perplexity", "configuration": "Decider V1 27B", "value": 598.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "liquid-d1", "model_name": "d1", "developer": "Liquid AI", "configuration": "d1", "value": 625.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "cloudflare-clef", "model_name": "Clef", "developer": "Cloudflare", "configuration": "Clef", "value": 713.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "google-gemini-3-1-flash-lite", "model_name": "Gemini 3.1 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.1 Flash-Lite (minimal reasoning)", "value": 792.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-5-4-mini", "model_name": "GPT-5.4 mini", "developer": "OpenAI", "configuration": "GPT-5.4 mini (no reasoning)", "value": 916.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "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": 917.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-5-4-nano", "model_name": "GPT-5.4 nano", "developer": "OpenAI", "configuration": "GPT-5.4 nano (no reasoning)", "value": 1061.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "google-gemini-3-5-flash", "model_name": "Gemini 3.5 Flash", "developer": "Google", "configuration": "Gemini 3.5 Flash (minimal reasoning)", "value": 1198.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 1241.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast (low reasoning)", "value": 1394.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "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 (low reasoning)", "value": 1440.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna (no reasoning)", "value": 1467.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol (no reasoning)", "value": 1573.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "google-gemini-3-6-flash", "model_name": "Gemini 3.6 Flash", "developer": "Google", "configuration": "Gemini 3.6 Flash (minimal reasoning)", "value": 1580.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "zai-glm-5-3", "model_name": "GLM-5.3", "developer": "Z.ai", "configuration": "GLM-5.3 (low reasoning)", "value": 1583.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5 (no reasoning)", "value": 1695.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast", "value": 1710.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash (low reasoning)", "value": 1813.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder (minimal reasoning)", "value": 1885.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol (low reasoning)", "value": 2197.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra (low reasoning)", "value": 2273.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3 (low reasoning)", "value": 2579.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra (low reasoning)", "value": 2799.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 2847.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 2920.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-14", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2 (low reasoning)", "value": 2958.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash (low reasoning)", "value": 2989.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 3009.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 3026.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview (low reasoning)", "value": 3090.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017 (low reasoning)", "value": 3334.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 3433.5, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "zai-glm-5-3", "model_name": "GLM-5.3", "developer": "Z.ai", "configuration": "GLM-5.3", "value": 3622.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3 (low reasoning)", "value": 4184.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3 (minimal reasoning)", "value": 4232.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.median_latency_ms", "metric_name": "BulletBench Bullet 60s: median move time", "unit": "milliseconds", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro (low reasoning)", "value": 5040.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "bytedance-seed-2-0-mini", "model_name": "Seed-2.0-Mini", "developer": "ByteDance", "configuration": "Seed-2.0-Mini (minimal reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "cloudflare-clef-flash", "model_name": "Clef Flash", "developer": "Cloudflare", "configuration": "Clef Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "cloudflare-clef", "model_name": "Clef", "developer": "Cloudflare", "configuration": "Clef", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "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": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3 (no reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "inception-mercury-decide", "model_name": "Mercury Decide", "developer": "Inception", "configuration": "Mercury Decide", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "liquid-d1", "model_name": "d1", "developer": "Liquid AI", "configuration": "d1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5 (no reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-luna-decisions", "model_name": "GPT-6 Luna Decisions", "developer": "OpenAI", "configuration": "GPT-6 Luna Decisions", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "thinkingmachines-inkling-small", "model_name": "Inkling Small", "developer": "Thinking Machines", "configuration": "Inkling Small (no reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "typesafe-jev-1-13", "model_name": "Jev 1.13", "developer": "TypeSafe", "configuration": "Jev 1.13", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "google-gemini-3-1-flash-lite", "model_name": "Gemini 3.1 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.1 Flash-Lite (minimal reasoning)", "value": 8.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 8.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "perplexity-pplx-decider-v1-27b", "model_name": "Decider V1 27B", "developer": "Perplexity", "configuration": "Decider V1 27B", "value": 8.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 25.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "google-gemini-3-5-flash", "model_name": "Gemini 3.5 Flash", "developer": "Google", "configuration": "Gemini 3.5 Flash (minimal reasoning)", "value": 25.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "mistral-mistral-small-2603", "model_name": "Mistral Small 4", "developer": "Mistral AI", "configuration": "Mistral Small 4 (no reasoning)", "value": 25.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast (low reasoning)", "value": 25.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5 (no reasoning)", "value": 37.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "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 (low reasoning)", "value": 37.5, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "google-gemini-3-6-flash", "model_name": "Gemini 3.6 Flash", "developer": "Google", "configuration": "Gemini 3.6 Flash (minimal reasoning)", "value": 41.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast", "value": 41.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash (low reasoning)", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-5-4-mini", "model_name": "GPT-5.4 mini", "developer": "OpenAI", "configuration": "GPT-5.4 mini (no reasoning)", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna (no reasoning)", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol (no reasoning)", "value": 50.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview (low reasoning)", "value": 58.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash", "value": 58.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder (minimal reasoning)", "value": 58.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2 (low reasoning)", "value": 62.5, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 62.5, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-5-4-nano", "model_name": "GPT-5.4 nano", "developer": "OpenAI", "configuration": "GPT-5.4 nano (no reasoning)", "value": 66.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash (low reasoning)", "value": 75.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol (low reasoning)", "value": 75.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra (low reasoning)", "value": 75.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra (low reasoning)", "value": 75.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "zai-glm-5-3", "model_name": "GLM-5.3", "developer": "Z.ai", "configuration": "GLM-5.3 (low reasoning)", "value": 83.33, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017 (low reasoning)", "value": 87.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 87.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-14", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3 (low reasoning)", "value": 87.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro (low reasoning)", "value": 87.5, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3 (minimal reasoning)", "value": 91.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3 (low reasoning)", "value": 91.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol", "value": 91.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.time_loss_rate", "metric_name": "BulletBench Bullet 60s: games lost on time", "unit": "% of games", "direction": "lower", "model": "zai-glm-5-3", "model_name": "GLM-5.3", "developer": "Z.ai", "configuration": "GLM-5.3", "value": 91.67, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2 (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5 (no reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "cloudflare-clef-flash", "model_name": "Clef Flash", "developer": "Cloudflare", "configuration": "Clef Flash", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "cloudflare-clef", "model_name": "Clef", "developer": "Cloudflare", "configuration": "Clef", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "deepseek-deepseek-v4-1-flash-1017", "model_name": "DeepSeek-V4.1-Flash-1017", "developer": "DeepSeek", "configuration": "DeepSeek-V4.1-Flash-1017 (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "google-gemini-3-1-flash-lite", "model_name": "Gemini 3.1 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.1 Flash-Lite (minimal reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "google-gemini-3-1-pro-preview", "model_name": "Gemini 3.1 Pro Preview", "developer": "Google", "configuration": "Gemini 3.1 Pro Preview (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "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": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "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": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "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": "2026-10-14", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3 (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3 (no reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "inception-mercury-decide", "model_name": "Mercury Decide", "developer": "Inception", "configuration": "Mercury Decide", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "liquid-d1", "model_name": "d1", "developer": "Liquid AI", "configuration": "d1", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3 (minimal reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder (minimal reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "moonshotai-kimi-k3-1-pro", "model_name": "Kimi K3.1 Pro", "developer": "Moonshot AI", "configuration": "Kimi K3.1 Pro (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3 (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-5-4-mini", "model_name": "GPT-5.4 mini", "developer": "OpenAI", "configuration": "GPT-5.4 mini (no reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "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": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "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": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-6-luna-decisions", "model_name": "GPT-6 Luna Decisions", "developer": "OpenAI", "configuration": "GPT-6 Luna Decisions", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna (no reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol (no reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "perplexity-pplx-decider-v1-27b", "model_name": "Decider V1 27B", "developer": "Perplexity", "configuration": "Decider V1 27B", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "typesafe-jev-1-13", "model_name": "Jev 1.13", "developer": "TypeSafe", "configuration": "Jev 1.13", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "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 (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "zai-glm-5-3", "model_name": "GLM-5.3", "developer": "Z.ai", "configuration": "GLM-5.3 (low reasoning)", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "zai-glm-5-3", "model_name": "GLM-5.3", "developer": "Z.ai", "configuration": "GLM-5.3", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "google-gemini-3-6-flash", "model_name": "Gemini 3.6 Flash", "developer": "Google", "configuration": "Gemini 3.6 Flash (minimal reasoning)", "value": 0.28, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "anthropic-claude-haiku-4-5", "model_name": "Claude Haiku 4.5", "developer": "Anthropic", "configuration": "Claude Haiku 4.5", "value": 0.31, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "bytedance-seed-2-0-mini", "model_name": "Seed-2.0-Mini", "developer": "ByteDance", "configuration": "Seed-2.0-Mini (minimal reasoning)", "value": 0.42, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5 (no reasoning)", "value": 0.43, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "google-gemini-3-5-flash", "model_name": "Gemini 3.5 Flash", "developer": "Google", "configuration": "Gemini 3.5 Flash (minimal reasoning)", "value": 0.46, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "mistral-mistral-small-2603", "model_name": "Mistral Small 4", "developer": "Mistral AI", "configuration": "Mistral Small 4 (no reasoning)", "value": 0.55, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "mistral-mistral-medium-3-5", "model_name": "Mistral Medium 3.5", "developer": "Mistral AI", "configuration": "Mistral Medium 3.5", "value": 0.79, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "openai-gpt-5-4-nano", "model_name": "GPT-5.4 nano", "developer": "OpenAI", "configuration": "GPT-5.4 nano (no reasoning)", "value": 3.06, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.invalid_move_rate", "metric_name": "BulletBench Bullet 60s: invalid moves", "unit": "% of moves", "direction": "lower", "model": "thinkingmachines-inkling-small", "model_name": "Inkling Small", "developer": "Thinking Machines", "configuration": "Inkling Small (no reasoning)", "value": 7.18, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "inception-mercury-decide", "model_name": "Mercury Decide", "developer": "Inception", "configuration": "Mercury Decide", "value": 0.0, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "liquid-d1", "model_name": "d1", "developer": "Liquid AI", "configuration": "d1", "value": 0.000733, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "perplexity-pplx-decider-v1-27b", "model_name": "Decider V1 27B", "developer": "Perplexity", "configuration": "Decider V1 27B", "value": 0.000964, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "typesafe-jev-1-13", "model_name": "Jev 1.13", "developer": "TypeSafe", "configuration": "Jev 1.13", "value": 0.001314, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "minimax-minimax-m3-1-flash", "model_name": "MiniMax M3.1 Flash", "developer": "MiniMax", "configuration": "MiniMax M3.1 Flash (low reasoning)", "value": 0.001584, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-luna", "model_name": "GPT-6 Luna", "developer": "OpenAI", "configuration": "GPT-6 Luna (no reasoning)", "value": 0.001624, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "anthropic-claude-haiku-5-5", "model_name": "Claude Haiku 5.5", "developer": "Anthropic", "configuration": "Claude Haiku 5.5 (no reasoning)", "value": 0.001915, "ci_low": null, "ci_high": null, "observed_on": "2026-10-08", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3 (no reasoning)", "value": 0.00195, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "cloudflare-clef-flash", "model_name": "Clef Flash", "developer": "Cloudflare", "configuration": "Clef Flash", "value": 0.002546, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-luna-decisions", "model_name": "GPT-6 Luna Decisions", "developer": "OpenAI", "configuration": "GPT-6 Luna Decisions", "value": 0.002591, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "bytedance-seed-2-0-mini", "model_name": "Seed-2.0-Mini", "developer": "ByteDance", "configuration": "Seed-2.0-Mini (minimal reasoning)", "value": 0.003021, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "inception-mercury-3", "model_name": "Mercury 3", "developer": "Inception", "configuration": "Mercury 3 (low reasoning)", "value": 0.003208, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "google-gemini-4-argon", "model_name": "Gemini 4 Argon", "developer": "Google", "configuration": "Gemini 4 Argon", "value": 0.003586, "ci_low": null, "ci_high": null, "observed_on": "2026-10-14", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "mistral-mistral-small-2603", "model_name": "Mistral Small 4", "developer": "Mistral AI", "configuration": "Mistral Small 4 (no reasoning)", "value": 0.003857, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "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 (low reasoning)", "value": 0.004732, "ci_low": null, "ci_high": null, "observed_on": "2026-10-19", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-5-4-nano", "model_name": "GPT-5.4 nano", "developer": "OpenAI", "configuration": "GPT-5.4 nano (no reasoning)", "value": 0.005024, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "cloudflare-clef", "model_name": "Clef", "developer": "Cloudflare", "configuration": "Clef", "value": 0.005627, "ci_low": null, "ci_high": null, "observed_on": "2026-10-07", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "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": 0.005796, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "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 (low reasoning)", "value": 0.006058, "ci_low": null, "ci_high": null, "observed_on": "2026-11-04", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "google-gemini-3-1-flash-lite", "model_name": "Gemini 3.1 Flash-Lite", "developer": "Google", "configuration": "Gemini 3.1 Flash-Lite (minimal reasoning)", "value": 0.006825, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "thinkingmachines-inkling-small", "model_name": "Inkling Small", "developer": "Thinking Machines", "configuration": "Inkling Small (no reasoning)", "value": 0.007487, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "google-gemini-3-6-flash", "model_name": "Gemini 3.6 Flash", "developer": "Google", "configuration": "Gemini 3.6 Flash (minimal reasoning)", "value": 0.011273, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "meta-muse-spark-1-3", "model_name": "Muse Spark 1.3", "developer": "Meta", "configuration": "Muse Spark 1.3 (minimal reasoning)", "value": 0.014236, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "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": 0.015378, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "zai-glm-5-3", "model_name": "GLM-5.3", "developer": "Z.ai", "configuration": "GLM-5.3", "value": 0.015582, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "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": 0.016258, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "google-gemini-3-8-flash", "model_name": "Gemini 3.8 Flash", "developer": "Google", "configuration": "Gemini 3.8 Flash (low reasoning)", "value": 0.016514, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-1-terra", "model_name": "GPT-6.1 Terra", "developer": "OpenAI", "configuration": "GPT-6.1 Terra (low reasoning)", "value": 0.017217, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "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 (low reasoning)", "value": 0.017541, "ci_low": null, "ci_high": null, "observed_on": "2026-10-27", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-5-4-mini", "model_name": "GPT-5.4 mini", "developer": "OpenAI", "configuration": "GPT-5.4 mini (no reasoning)", "value": 0.020513, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "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": 0.020984, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "zai-glm-5-3", "model_name": "GLM-5.3", "developer": "Z.ai", "configuration": "GLM-5.3 (low reasoning)", "value": 0.021951, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol (low reasoning)", "value": 0.024288, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "google-gemini-3-5-flash", "model_name": "Gemini 3.5 Flash", "developer": "Google", "configuration": "Gemini 3.5 Flash (minimal reasoning)", "value": 0.028933, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-sol", "model_name": "GPT-6 Sol", "developer": "OpenAI", "configuration": "GPT-6 Sol (no reasoning)", "value": 0.029717, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "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 (no reasoning)", "value": 0.035407, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "none", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "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": 0.041972, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "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 (low reasoning)", "value": 0.044593, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "Gemini API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "moonshotai-kimi-k3", "model_name": "Kimi K3", "developer": "Moonshot AI", "configuration": "Kimi K3 (low reasoning)", "value": 0.053958, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "meta-muse-wonder", "model_name": "Muse Wonder", "developer": "Meta", "configuration": "Muse Wonder (minimal reasoning)", "value": 0.066301, "ci_low": null, "ci_high": null, "observed_on": "2026-10-12", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "minimal", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "anthropic-claude-fable-5-2", "model_name": "Claude Fable 5.2", "developer": "Anthropic", "configuration": "Claude Fable 5.2 (low reasoning)", "value": 0.079975, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra", "value": 0.094311, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "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": 0.094843, "ci_low": null, "ci_high": null, "observed_on": "2026-11-10", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-astra", "model_name": "GPT-6 Astra", "developer": "OpenAI", "configuration": "GPT-6 Astra (low reasoning)", "value": 0.103691, "ci_low": null, "ci_high": null, "observed_on": "2026-10-01", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenRouter", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast (low reasoning)", "value": 0.336607, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "low", "served_by": "OpenAI API (direct)", "judge": ""}, {"benchmark": "bulletbench", "metric": "60.cost_per_game", "metric_name": "BulletBench Bullet 60s: cost per game", "unit": "US dollars", "direction": "lower", "model": "openai-gpt-6-1-sol", "model_name": "GPT-6.1 Sol", "developer": "OpenAI", "configuration": "GPT-6.1 Sol Ultrafast", "value": 0.35722, "ci_low": null, "ci_high": null, "observed_on": "2026-10-09", "tasks": "", "products": "", "repeats": "", "reasoning_effort": "provider default", "served_by": "OpenAI API (direct)", "judge": ""}]}