Best for reasoning
Ranked by GPQA Diamond - graduate-level science questions written to be hard to look up.
Ranked by GPQA Diamond - graduate-level science questions written to be hard to look up. Source: Epoch AI. Variants (reasoning effort, batch) are folded into one row.
| # | Model | Capability | Reasoning | Coding | Context | $/M | Weights | Released |
|---|---|---|---|---|---|---|---|---|
| 101 | GPT-4.5 Preview (Feb 2025)OpenAI | — | 68.7 | — | — | — | Closed | 27 Feb 2025 |
| 102 | DeepSeek-V3 (Mar 2025)DeepSeek | 135.9 | 67.6 | — | — | — | Open | 24 Mar 2025 |
| 103 | Llama 4 Maverick (FP8)Meta | — | 67.0 | — | — | — | Open | 5 Apr 2025 |
| 104 | GPT-4.1OpenAI | 136.8 | 66.9 | — | 1.05M | $3.50 | Closed | 14 Apr 2025 |
| 105 | GPT-4.1 MiniOpenAI | 135.0 | 65.8 | — | 1.05M | $0.70 | Closed | 14 Apr 2025 |
| 106 | Qwen3 32BAlibaba (Qwen) | 138.5 | 65.7 | — | 131K | $0.13 | Open | 29 Apr 2025 |
| 107 | Gemini 2.0 Pro Exp (Feb 2025)Google | — | 65.7 | — | — | — | Closed | 5 Feb 2025 |
| 108 | QWQ-PlusAlibaba (Qwen) | — | 65.4 | — | — | — | Closed | 8 Apr 2025 |
| 109 | QwQ-32BAlibaba (Qwen) | 137.6 | 65.3 | — | — | — | Open | 5 Mar 2025 |
| 110 | DeepSeek-R1-Distill-Qwen-32BDeepSeek | 137.4 | 64.1 | — | — | — | Open | 20 Jan 2025 |
| 111 | Gemini 2.0 Flash (Feb 2025)Google | 134.7 | 64.1 | — | — | — | Closed | 5 Feb 2025 |
| 112 | Qwen3 14BAlibaba (Qwen) | 138.2 | 63.8 | — | 131K | $0.15 | Open | 29 Apr 2025 |
| 113 | o1-miniOpenAI · +1 variant | 135.8 | 62.4 | — | — | — | Closed | 12 Sep 2024 |
| 114 | Qwen3 30B A3BAlibaba (Qwen) | 136.2 | 61.7 | — | 131K | $0.23 | Open | 29 Apr 2025 |
| 115 | gpt-oss-20bOpenAI | 137.8 | 60.8 | — | 131K | $0.036 | Open | 5 Aug 2025 |
| 116 | GLM 4.7 FlashZ.ai (Zhipu AI) | — | 60.5 | — | 200K | $0.15 | Open | 19 Jan 2026 |
| 117 | Mistral Medium 3Mistral AI | 134.1 | 59.5 | — | 131K | $0.80 | Closed | 7 May 2025 |
| 118 | Gemini 1.5 Pro (Sept 2024)Google | 131.7 | 57.2 | — | — | — | Closed | 24 Sep 2024 |
| 119 | Gemini 2.0 Flash Thinking ExpGoogle | — | 57.1 | — | — | — | Closed | 21 Jan 2025 |
| 120 | Qwen3 8BAlibaba (Qwen) | 136.2 | 56.8 | — | 131K | $0.20 | Open | 28 Apr 2025 |
| 121 | DeepSeek V3DeepSeek | 132.4 | 56.5 | — | 164K | $0.45 | Open | 26 Dec 2024 |
| 122 | Qwen2.5-MaxAlibaba (Qwen) | 132.5 | 56.1 | — | — | — | Closed | 25 Jan 2025 |
| 123 | Magistral Small 1.0Mistral AI | 133.2 | 56.1 | — | — | — | Open | 10 Jun 2025 |
| 124 | Phi 4Microsoft | 130.4 | 56.1 | — | 16K | $0.087 | Open | 12 Dec 2024 |
| 125 | DeepSeek-R1-Distill-Llama-70BDeepSeek | — | 55.7 | — | — | — | Open | 20 Jan 2025 |
| 126 | Qwen3-30B-A3B-Instruct (Jul 2025)Alibaba (Qwen) | 137.4 | 55.6 | — | — | — | Open | 29 Jul 2025 |
| 127 | Claude 3.5 Sonnet (Oct 2024)Anthropic | — | 55.3 | — | — | — | Closed | 22 Oct 2024 |
| 128 | Claude 3.5 Sonnet (Jun 2024)Anthropic | — | 54.0 | — | — | — | Closed | 20 Jun 2024 |
| 129 | Grok-2 (Dec 2024)xAI | 130.5 | 53.8 | — | — | — | Closed | 12 Dec 2024 |
| 130 | Qwen3-4BAlibaba (Qwen) | — | 52.3 | — | — | — | Open | 29 Apr 2025 |
| 131 | Llama 4 ScoutMeta | 129.6 | 51.8 | — | 1.31M | $0.15 | Open | 5 Apr 2025 |
| 132 | Mistral Large 2 (Nov 2024)Mistral AI | 128.5 | 51.3 | — | — | — | Open | 18 Nov 2024 |
| 133 | Llama 3.1-405BMeta | 128.8 | 50.9 | — | — | — | Open | 23 Jul 2024 |
| 134 | o1-previewOpenAI | 134.8 | 50.3 | — | — | — | Closed | 12 Sep 2024 |
| 135 | GPT-4o (Aug 2024)OpenAI | 128.8 | 49.2 | — | — | — | Closed | 6 Aug 2024 |
| 136 | Qwen2.5-72BAlibaba (Qwen) | 129.0 | 49.1 | — | — | — | Open | 19 Sep 2024 |
| 137 | Mistral Small 3.2Mistral AI | 131.7 | 49.1 | — | — | — | Open | 20 Jun 2025 |
| 138 | Mistral Large 2 (Jul 2024)Mistral AI | 127.5 | 49.0 | — | — | — | Open | 24 Jul 2024 |
| 139 | GPT-4.1 NanoOpenAI | 129.6 | 48.9 | — | 1.05M | $0.17 | Closed | 14 Apr 2025 |
| 140 | GPT-4o (May 2024)OpenAI | 129.0 | 48.9 | — | — | — | Closed | 13 May 2024 |
| 141 | Qwen Plus (Jan 2025)Alibaba (Qwen) | — | 48.1 | — | — | — | Closed | 25 Jan 2025 |
| 142 | GPT-4o (Nov 2024)OpenAI | 128.8 | 47.9 | — | — | — | Closed | 20 Nov 2024 |
| 143 | Gemma 3 27BGoogle | 130.0 | 47.7 | — | 131K | $0.17 | Open | 12 Mar 2025 |
| 144 | Magistral Small 1.2Mistral AI | 131.4 | 47.6 | — | — | — | Open | 18 Sep 2025 |
| 145 | Mistral Small 3.1Mistral AI | 127.5 | 47.5 | — | — | — | Open | 17 Mar 2025 |
| 146 | Llama 3.3 70BMeta | 127.3 | 47.4 | — | — | — | Open | 6 Dec 2024 |
| 147 | Gemini 1.5 Flash (Sep 2024)Google | 129.4 | 47.3 | — | — | — | Closed | 24 Sep 2024 |
| 148 | Mistral Small 3Mistral AI | 127.1 | 47.3 | — | 33K | $0.058 | Open | 30 Jan 2025 |
| 149 | Claude 3 OpusAnthropic | 126.9 | 47.2 | — | — | — | Closed | 29 Feb 2024 |
| 150 | GPT-4 Turbo (Apr 2024)OpenAI | 127.3 | 46.6 | — | — | — | Closed | 9 Apr 2024 |
— means no published score from that source yet · click a model for every benchmark with its source
New - awaiting independent evaluation
All new models →- Claude Sonnet 5.5Anthropic · 28 Sep 2026 · AA 56 · 1M · $4.00/M
- Qwen3.8 Max PrimeAlibaba (Qwen) · 23 Sep 2026 · 1M · $6.00/M
- GPT-6 SolOpenAI · 22 Sep 2026 · AA 47.5 · 1.05M · $4.00/M
- GPT-6 LunaOpenAI · 22 Sep 2026 · AA 37.3 · 1.05M · $0.20/M
- GPT-6 Sol ProOpenAI · 22 Sep 2026 · 1.05M · $4.00/M
- GPT-6 Luna ProOpenAI · 22 Sep 2026 · 1.05M · $0.20/M
- Claude Opus 5.5Anthropic · 22 Sep 2026 · AA 57.6 · 1M · $8.00/M
- Grok 4.7xAI · 21 Sep 2026 · AA 46.4 · 500K · $3.00/M
Not ranked until an independent evaluator (Epoch AI) publishes scores. We don't use vendor-reported benchmark claims for rankings.
Best AI for coding →
Models ranked on DeepSWE
Best AI for reasoning →
Ranked on GPQA Diamond
Best AI for math →
Ranked on competition math
Best open-weight models →
Weights you can run yourself
Cheapest capable models →
Lowest blended API price
Best AI coding tools →
Apps & agents by popularity
Best AI image generators →
Apps & models
Best AI for research →
Answer engines & deep research