AI models compared
584 models with published results from our own benchmarks and licensed sources, strongest first: by our overall score, then the Artificial Analysis Intelligence Index. Open a model for every result, with its source and date. 225 of them are on sale through OpenRouter, with prices on the API pricing page; launches are on the release tracker.
Sorted by Spring Prompt overall, lowest first; models without a value come last. Select a column heading to sort by it.
| Model | Spring Prompt overall | Intelligence Index | Price, blended | Released | Context | Sources |
|---|---|---|---|---|---|---|
| Qwen3.5-FlashAlibaba | – | – | $0.11 | – | 1,000k | 5 |
| Mistral Large 2.1 (2411)Mistral AI · not on sale | – | – | – | 18 Nov 2024 | – | 4 |
| Claude 3 OpusAnthropic · not on sale | – | – | – | 4 Mar 2024 | – | 3 |
| Claude Opus 4Anthropic · not on sale | – | – | – | 22 May 2025 | – | 3 |
| Command R+ (08-2024)Cohere | – | – | $4.38 | – | 128k | 3 |
| Grok 4 (0709)xAI · not on sale | – | – | – | – | – | 3 |
| Ministral 8B (2410)Mistral AI · not on sale | – | – | – | 16 Oct 2024 | – | 3 |
| Qwen3.5-Plus (2026-02-15)Alibaba | – | – | $0.58 | 16 Feb 2026 | 1,000k | 3 |
| Trinity Large PreviewArcee AI · not on sale | – | – | – | – | – | 3 |
| Aya Expanse 32BCohere · not on sale | – | – | – | – | – | 2 |
| Aya Expanse 8BCohere · not on sale | – | – | – | – | – | 2 |
| ChatGPT-4o (2025-03-26)OpenAI · not on sale | – | – | – | – | – | 2 |
| Claude 3 HaikuAnthropic · not on sale | – | – | – | 13 Mar 2024 | – | 2 |
| Claude 3.7 SonnetAnthropic · not on sale | – | – | – | 24 Feb 2025 | – | 2 |
| Command R (08-2024)Cohere | – | – | $0.26 | 30 Aug 2024 | 128k | 2 |
| Gemma 2 27BGoogle · open weights | – | – | $0.65 | 24 Jun 2024 | 8k | 2 |
| Gemma 2 2BGoogle · not on sale | – | – | – | – | – | 2 |
| Gemma 2 9BGoogle · not on sale | – | – | – | – | – | 2 |
| GPT-5.3 ChatOpenAI · not on sale | – | – | – | – | – | 2 |
| GPT-5.4 ProOpenAI | – | – | $67.50 | 5 Mar 2026 | 1,050k | 2 |
| Granite 3.1 8B InstructIBM · not on sale | – | – | – | – | – | 2 |
| Granite 4.0 H SmallIBM · not on sale | – | – | – | – | – | 2 |
| Grok 3xAI · not on sale | – | – | – | 17 Feb 2025 | – | 2 |
| Grok 4 FastxAI · not on sale | – | – | – | 19 Sep 2025 | – | 2 |
| Grok 4.1 FastxAI · not on sale | – | – | – | – | – | 2 |
| Grok 4.20 Beta (0309, reasoning)xAI · not on sale | – | – | – | – | – | 2 |
| Grok 4.20 Multi-Agent Beta (0309)xAI · not on sale | – | – | – | – | – | 2 |
| Jamba Large 1.7AI21 Labs · not on sale | – | – | – | – | – | 2 |
| Jamba Mini 1.7AI21 Labs · not on sale | – | – | – | – | – | 2 |
| Llama 2 70B ChatMeta · not on sale | – | – | – | – | – | 2 |
| Llama 3.1 Nemotron Ultra 253B v1NVIDIA · not on sale | – | – | – | – | – | 2 |
| Llama 4 Maverick 17B 128E Instruct (FP8)Meta · not on sale | – | – | – | – | – | 2 |
| MiMo-V2-FlashXiaomi · not on sale | – | – | – | 16 Dec 2025 | – | 2 |
| Mistral NemoMistral AI · open weights | – | – | $0.02 | 18 Jul 2024 | 131k | 2 |
| Mixtral 8x7B Instruct v0.1Mistral AI · not on sale | – | – | – | – | – | 2 |
| Nanbeige4-3B-Thinking (2511)Nanbeige · not on sale | – | – | – | – | – | 2 |
| Olmo 3 32B ThinkAi2 · not on sale | – | – | – | – | – | 2 |
| Qwen3-0.6BAlibaba · not on sale | – | – | – | – | – | 2 |
| Qwen3-1.7BAlibaba · not on sale | – | – | – | – | – | 2 |
| Qwen3-4BAlibaba · not on sale | – | – | – | – | – | 2 |
| QwQ-32BAlibaba · not on sale | – | – | – | – | – | 2 |
| Snowflake Arctic InstructSnowflake · not on sale | – | – | – | – | – | 2 |
| Solar 10.7B Instruct v1.0Upstage · not on sale | – | – | – | – | – | 2 |
| Alpaca 13BStanford · not on sale | – | – | – | – | – | 1 |
| Apriel Nemotron 15B ThinkerServiceNow · not on sale | – | – | – | – | – | 1 |
| Arch-Agent-1.5BKatanemo · not on sale | – | – | – | – | – | 1 |
| Arch-Agent-32BKatanemo · not on sale | – | – | – | – | – | 1 |
| Arch-Agent-3BKatanemo · not on sale | – | – | – | – | – | 1 |
| Athene 70B (0725)Nexusflow · not on sale | – | – | – | – | – | 1 |
| Athene-V2-ChatNexusflow · not on sale | – | – | – | – | – | 1 |
Spring Prompt overall is our score across every source, out of 100 (how it works). The Intelligence Index is Artificial Analysis's, at each model's best setting. Price is the developer's list price, or the typical OpenRouter host, for three input tokens to one output, per million tokens.
Want two models side by side? Compare models.