Best for math
Ranked by Epoch's mock AIME 2024-2025 (competition mathematics).
Ranked by Epoch's mock AIME 2024-2025 (competition mathematics). Source: Epoch AI. Variants (reasoning effort, batch) are folded into one row.
| # | Model | Capability | Reasoning | Coding | Context | $/M | Weights | Released |
|---|---|---|---|---|---|---|---|---|
| 1 | GPT-6 AstraNEWOpenAI | 166.6 | 95.8 | 73.2 | 1.05M | $20.00 | Closed | 3 Sep 2026 |
| 2 | Claude Fable 5.1NEWAnthropic | 165.0 | — | — | 1M | $20.00 | Closed | 1 Sep 2026 |
| 3 | Claude Fable 5Anthropic · +1 variant | 163.6 | 85.9 | 69.7 | 1M | $20.00 | Closed | 9 Jun 2026 |
| 4 | GPT-5.6 SolOpenAI · +2 variants | 162.0 | 93.5 | 72.7 | 1.05M | $4.00 | Closed | 9 Jul 2026 |
| 5 | GPT-5.5xhighOpenAI · +2 variants | — | 94.0 | 67.0 | — | — | Closed | 23 Apr 2026 |
| 6 | GPT-5.5 ProxhighOpenAI | — | 93.9 | — | — | — | Closed | 23 Apr 2026 |
| 7 | Qwen3.8 Max (0902)NEWxhighAlibaba (Qwen) | — | 92.3 | — | — | — | Closed | 1 Sep 2026 |
| 8 | GPT-5.6 TerraOpenAI · +2 variants | 159.3 | 93.3 | 69.6 | 1.05M | $4.50 | Closed | 9 Jul 2026 |
| 9 | Qwen3.8 MaxxhighAlibaba (Qwen) | — | 92.7 | 57.5 | — | — | Closed | 2 Aug 2026 |
| 10 | Grok 4.6xhighxAI · +1 variant | — | 93.2 | 66.7 | — | — | Closed | 12 Aug 2026 |
| 11 | Muse Spark 1.3NEWxhighMeta | — | — | — | — | — | Closed | 2 Sep 2026 |
| 12 | Claude Opus 5Anthropic · +1 variant | 162.7 | 93.9 | 73.6 | 1M | $10.00 | Closed | 24 Jul 2026 |
| 13 | Gemini 3.8 FlashNEWGoogle | 157.1 | 95.4 | 73.8 | 1.05M | $1.50 | Closed | 2 Sep 2026 |
| 14 | DeepSeek V4 Pro 0813DeepSeek | 155.4 | 91.7 | — | 1.05M | $1.41 | Open | 13 Aug 2026 |
| 15 | Claude Opus 4.8Anthropic | 158.3 | 91.0 | 59.0 | 1M | $10.00 | Closed | 28 May 2026 |
| 16 | GPT-5.6 LunaOpenAI · +2 variants | 156.3 | 91.6 | 67.2 | 1.05M | $0.45 | Closed | 9 Jul 2026 |
| 17 | Claude Opus 4.7xhighAnthropic · +1 variant | — | 90.2 | — | — | — | Closed | 16 Apr 2026 |
| 18 | GPT-5.4OpenAI · +4 variants | 156.9 | 89.9 | — | 1.05M | $5.63 | Closed | 5 Mar 2026 |
| 19 | Grok 4.5xAI | 154.0 | 93.4 | 53.8 | 500K | $3.00 | Closed | 8 Jul 2026 |
| 20 | Gemini 3.7 FlashGoogle | 157.7 | 94.8 | 65.3 | 1.05M | $1.50 | Closed | 13 Aug 2026 |
| 21 | Kimi K3Moonshot AI · +1 variant | 157.7 | 93.1 | 68.5 | 1.05M | $6.00 | Open | 16 Jul 2026 |
| 22 | DeepSeek v4 ProhighDeepSeek · +1 variant | — | 90.9 | — | — | — | Open | 24 Apr 2026 |
| 23 | GPT-5.2OpenAI · +4 variants | 153.4 | 88.2 | — | 400K | $4.81 | Closed | 11 Dec 2025 |
| 24 | Kimi K2.6Moonshot AI | 151.0 | 90.8 | — | 262K | $1.34 | Open | 20 Apr 2026 |
| 25 | Gemini 3.1 Pro PreviewGoogle | — | 94.4 | 11.7 | 1.05M | $4.50 | — | 19 Feb 2026 |
| 26 | Gemini 3.5 FlashGoogle | 154.6 | 92.8 | 37.4 | 1.05M | $3.38 | Closed | 19 May 2026 |
| 27 | Qwen3.7 MaxAlibaba (Qwen) | 153.7 | 90.9 | — | 1M | $2.21 | Closed | 19 May 2026 |
| 28 | Kimi K2.7 CodeMoonshot AI | 150.0 | 87.9 | 30.5 | 262K | $1.32 | Open | 12 Jun 2026 |
| 29 | Gemini 3 Flash PreviewGoogle | — | 89.4 | — | 1.05M | $1.13 | — | 17 Dec 2025 |
| 30 | Claude Sonnet 5xhighAnthropic · +1 variant | — | 90.5 | 49.7 | — | — | Closed | 30 Jun 2026 |
| 31 | DeepSeek V4 Flash 0731DeepSeek | 154.5 | 91.0 | — | 1.31M | $0.093 | Open | 31 Jul 2026 |
| 32 | Claude Opus 4.664k thinkingAnthropic · +2 variants | — | 88.8 | — | — | — | Closed | 5 Feb 2026 |
| 33 | Gemini 3.6 FlashGoogle | 154.4 | 94.1 | 46.7 | 1.05M | $1.50 | Closed | 21 Jul 2026 |
| 34 | GLM 5.3 FlashZ.ai (Zhipu AI) | 151.9 | 90.2 | 63.4 | 1.31M | $0.24 | Open | 20 Aug 2026 |
| 35 | GLM 5.1Z.ai (Zhipu AI) | 149.9 | 89.9 | — | 205K | $2.15 | Open | 7 Apr 2026 |
| 36 | Grok 4.3 BetaxAI | 149.1 | 88.8 | — | — | — | Closed | 17 Apr 2026 |
| 37 | Qwen3.7 PlusAlibaba (Qwen) | 147.4 | 87.9 | — | 1M | $0.56 | Closed | 2 Jun 2026 |
| 38 | Qwen 3.6 Plus (2026-04-02)Alibaba (Qwen) | — | 88.4 | — | — | — | Closed | 31 Mar 2026 |
| 39 | Grok 4.20xAI | 152.0 | 89.3 | — | 2M | $1.56 | Closed | 17 Feb 2026 |
| 40 | Kimi K2.5 (Fireworks)Moonshot AI | — | 87.6 | — | — | — | Open | 27 Jan 2026 |
| 41 | GPT-5OpenAI · +2 variants | 150.0 | 86.2 | — | 400K | $3.44 | Closed | 7 Aug 2025 |
| 42 | Gemini 3 Pro PreviewGoogle | — | 92.6 | — | — | — | Closed | 18 Nov 2025 |
| 43 | GLM 5.3Z.ai (Zhipu AI) | 155.6 | 90.9 | 69.0 | 1.31M | $2.15 | Open | 14 Aug 2026 |
| 44 | Qwen 3.6 Max (Preview)Alibaba (Qwen) | 149.3 | 87.4 | — | — | — | Closed | 20 Apr 2026 |
| 45 | Qwen3.6 27BAlibaba (Qwen) | 146.5 | 85.9 | — | 262K | $1.04 | Open | 22 Apr 2026 |
| 46 | Inkling SmallxhighThinking Machines Lab | — | 88.5 | — | — | — | Open | 15 Jul 2026 |
| 47 | Muse SparkMeta | 152.1 | 89.8 | — | — | — | Closed | 8 Apr 2026 |
| 48 | Qwen3.5 397B A17BAlibaba (Qwen) | 146.6 | 86.4 | — | 262K | $1.29 | Open | 13 Feb 2026 |
| 49 | gpt-oss-120bOpenAI | 139.9 | 75.8 | — | 131K | $0.070 | Open | 5 Aug 2025 |
| 50 | GPT-5.4 minixhighOpenAI · +2 variants | — | 86.9 | — | — | — | Closed | 17 Mar 2026 |
— 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