Updated daily at 05:00 UTC
Benchmark
OTIS Mock AIME 2024-2025
Measures: 45 competition problems from three OTIS Mock AIME exams with integer answers from 0 to 999, harder than MATH Level 5 and easier than FrontierMath.
- Maintainer
- Epoch AI (problems by OTIS students)
- Introduced
- 2024
- Status
- Active
- Saturation
- Saturated
- Typical provenance
- Independent
- In scoring set
- No
Leaderboard
Independent and third-party results only.
| Model | Accuracy | Provenance | Source | Evaluated |
|---|---|---|---|---|
| Claude Fable 5 Anthropic · High effort | 100.0% | Independent | Epoch AI | 6 Aug 2026 |
| Claude Fable 5.1 Anthropic · Max effort | 100.0% | Independent | Epoch AI | 1 Sep 2026 |
| Claude Opus 5.5 Anthropic · Max effort | 100.0% | Independent | Epoch AI | 22 Sep 2026 |
| GPT-5.6 Sol OpenAI · Max effort | 100.0% | Independent | Epoch AI | 9 Jul 2026 |
| GPT-6 Astra OpenAI · Max effort | 100.0% | Independent | Epoch AI | 30 Aug 2026 |
| Qwen3.8-Max Alibaba Group · Xhigh effort | 100.0% | Independent | Epoch AI | 2 Sep 2026 |
| Claude Fable 5 Anthropic · Max effort | 99.7% | Independent | Epoch AI | 10 Jun 2026 |
| GPT-5.6 Terra OpenAI · Max effort | 99.7% | Independent | Epoch AI | 9 Jul 2026 |
| Qwen3.8-Max Alibaba Group · Xhigh effort | 99.4% | Independent | Epoch AI | 4 Aug 2026 |
| Grok 4.6 SpaceXAI (xAI) · Xhigh effort | 99.2% | Independent | Epoch AI | 14 Aug 2026 |
| Muse Spark 1.3 Meta · Xhigh effort | 99.2% | Independent | Epoch AI | 17 Sep 2026 |
| Claude Opus 5 Anthropic · Max effort | 98.9% | Independent | Epoch AI | 24 Jul 2026 |
| Gemini 3.8 Flash Google DeepMind · High effort | 98.9% | Independent | Epoch AI | 2 Sep 2026 |
| DeepSeek-V4-Pro DeepSeek · Max effort | 98.6% | Independent | Epoch AI | 18 Aug 2026 |
| Claude Opus 4.8 Anthropic · Max effort | 98.3% | Independent | Epoch AI | 7 Jun 2026 |
| GPT-5.6 Luna OpenAI · Max effort | 98.3% | Independent | Epoch AI | 9 Jul 2026 |
| Claude Opus 4.7 Anthropic · Xhigh effort | 97.8% | Independent | Epoch AI | 17 Apr 2026 |
| Claude Fable 5 Anthropic · Low effort | 97.8% | Independent | Epoch AI | 6 Aug 2026 |
| Claude Opus 4.8 Anthropic · Low effort | 97.8% | Independent | Epoch AI | 6 Aug 2026 |
| Claude Opus 5 | 97.8% | Independent | Epoch AI | 6 Aug 2026 |
| GPT-5.4 OpenAI · High effort | 97.8% | Independent | Epoch AI | 15 Jul 2026 |
| Grok 4.5 SpaceXAI (xAI) · High effort | 97.8% | Independent | Epoch AI | 8 Jul 2026 |
| Grok 4.6 SpaceXAI (xAI) · High effort | 97.8% | Independent | Epoch AI | 12 Aug 2026 |
| Gemini 3.7 Flash Google DeepMind · High effort | 97.2% | Independent | Epoch AI | 14 Aug 2026 |
| Kimi K3 Moonshot AI · Max effort | 97.2% | Independent | Epoch AI | 16 Jul 2026 |
| DeepSeek-V4-Pro DeepSeek · Max effort | 96.7% | Independent | Epoch AI | 17 Jun 2026 |
| GPT-5.2 OpenAI · High effort | 96.1% | Independent | Epoch AI | 11 Dec 2025 |
| Kimi K2.6 | 96.1% | Independent | Epoch AI | 2 May 2026 |
| GPT-5.2 OpenAI · Xhigh effort | 96.1% | Independent | Epoch AI | 13 Dec 2025 |
| Gemini 3.1 Pro | 95.6% | Independent | Epoch AI | 20 Feb 2026 |
| DeepSeek-V4-Pro DeepSeek · High effort | 95.6% | Independent | Epoch AI | 6 Aug 2026 |
| Gemini 3 Flash Google DeepMind · High effort | 95.6% | Independent | Epoch AI | 6 Aug 2026 |
| Gemini 3.1 Pro Google DeepMind · High effort | 95.6% | Independent | Epoch AI | 6 Aug 2026 |
| Gemini 3.5 Flash Google DeepMind · High effort | 95.6% | Independent | Epoch AI | 25 May 2026 |
| GPT-5.4 OpenAI · Medium effort | 95.6% | Independent | Epoch AI | 15 Jul 2026 |
| GPT-5.6 Sol OpenAI · Low effort | 95.6% | Independent | Epoch AI | 7 Aug 2026 |
| Kimi K2.7 Code | 95.6% | Independent | Epoch AI | 7 Aug 2026 |
| Qwen3.7-Max | 95.6% | Independent | Epoch AI | 7 Aug 2026 |
| GPT-5.4 OpenAI · Xhigh effort | 95.3% | Independent | Epoch AI | 6 Mar 2026 |
| Claude Sonnet 5 Anthropic · Xhigh effort | 94.7% | Independent | Epoch AI | 1 Jul 2026 |
| Claude Opus 4.6 Anthropic · 64K effort | 94.4% | Independent | Epoch AI | 6 Feb 2026 |
| DeepSeek-V4-Flash DeepSeek · Max effort | 94.4% | Independent | Epoch AI | 2 Aug 2026 |
| Gemini 3.6 Flash Google DeepMind · High effort | 94.2% | Independent | Epoch AI | 2 Aug 2026 |
| GLM-5.3-Flash Z.ai (Zhipu AI) · Max effort | 93.9% | Independent | Epoch AI | 26 Aug 2026 |
| GPT-5.2 OpenAI · Medium effort | 93.9% | Independent | Epoch AI | 11 Dec 2025 |
| Claude Opus 5 Anthropic · Low effort | 93.3% | Independent | Epoch AI | 6 Aug 2026 |
| GLM-5.1 | 93.3% | Independent | Epoch AI | 10 Aug 2026 |
| Grok 4.3 SpaceXAI (xAI) · High effort | 93.3% | Independent | Epoch AI | 17 Jun 2026 |
| Kimi K3 Moonshot AI · High effort | 93.3% | Independent | Epoch AI | 7 Aug 2026 |
| Qwen3.6-Plus | 93.3% | Independent | Epoch AI | 7 Aug 2026 |
| Qwen3.7-Plus | 93.3% | Independent | Epoch AI | 7 Aug 2026 |
| Claude Opus 4.6 Anthropic · 32K effort | 93.1% | Independent | Epoch AI | 6 Feb 2026 |
| Gemini 3 Flash | 92.8% | Independent | Epoch AI | 17 Dec 2025 |
| Grok 4.20 | 92.2% | Independent | Epoch AI | 13 Jul 2026 |
| Gemini 3 Pro | 91.4% | Independent | Epoch AI | 19 Nov 2025 |
| GPT-5 OpenAI · High effort | 91.4% | Independent | Epoch AI | 29 Oct 2025 |
| Claude Opus 4.6 Anthropic · Max effort | 91.1% | Independent | Epoch AI | 6 Aug 2026 |
| GLM-5.3 Z.ai (Zhipu AI) · Max effort | 91.1% | Independent | Epoch AI | 24 Aug 2026 |
| Qwen3.6-27B | 91.1% | Independent | Epoch AI | 7 Aug 2026 |
| Qwen3.6-Max-Preview | 91.1% | Independent | Epoch AI | 7 Aug 2026 |
| Inkling Small Thinking Machines Lab · Xhigh effort | 90.0% | Independent | Epoch AI | 14 Aug 2026 |
| Gemini 3.5 Flash Google DeepMind · Low effort | 88.9% | Independent | Epoch AI | 6 Aug 2026 |
| GPT-5.4 mini OpenAI · Xhigh effort | 88.9% | Independent | Epoch AI | 7 Aug 2026 |
| GPT-5.6 Terra OpenAI · Low effort | 88.9% | Independent | Epoch AI | 7 Aug 2026 |
| gpt-oss-120b OpenAI · High effort | 88.9% | Independent | Epoch AI | 11 Dec 2025 |
| Inkling Thinking Machines Lab · Xhigh effort | 88.9% | Independent | Epoch AI | 5 Aug 2026 |
| Qwen3.5-397B-A17B | 88.9% | Independent | Epoch AI | 7 Aug 2026 |
| GPT-5.1 OpenAI · High effort | 88.6% | Independent | Epoch AI | 13 Nov 2025 |
| DeepSeek-V3.2 | 87.8% | Independent | Epoch AI | 16 Dec 2025 |
| GPT-5.4 nano OpenAI · High effort | 87.8% | Independent | Epoch AI | 14 Apr 2026 |
| GPT-5 OpenAI · Medium effort | 87.2% | Independent | Epoch AI | 7 Aug 2025 |
| GPT-5.4 mini OpenAI · High effort | 87.2% | Independent | Epoch AI | 15 Apr 2026 |
| Claude Opus 4.7 Anthropic · Max effort | 86.7% | Independent | Epoch AI | 6 Aug 2026 |
| GPT-5 mini OpenAI · High effort | 86.7% | Independent | Epoch AI | 30 Oct 2025 |
| Qwen3-235B-A22B-Thinking-2507 | 86.7% | Independent | Epoch AI | 10 Dec 2025 |
| Qwen3.5-Plus | 86.7% | Independent | Epoch AI | 7 Aug 2026 |
| Qwen3.6-35B-A3B | 86.7% | Independent | Epoch AI | 7 Aug 2026 |
| Qwen3.7-Flash | 86.7% | Independent | Epoch AI | 7 Aug 2026 |
| GLM-5.2 Z.ai (Zhipu AI) · Max effort | 86.4% | Independent | Epoch AI | 25 Jun 2026 |
| Claude Opus 4.5 Anthropic · 32K effort | 86.1% | Independent | Epoch AI | 24 Nov 2025 |
| Claude Sonnet 4.6 Anthropic · 32K effort | 85.8% | Independent | Epoch AI | 20 Feb 2026 |
| GPT-5.1 OpenAI · Medium effort | 85.6% | Independent | Epoch AI | 17 Nov 2025 |
| Gemini 2.5 Pro | 84.7% | Independent | Epoch AI | 16 Nov 2025 |
| Claude Opus 4.8 Anthropic · None effort | 84.4% | Independent | Epoch AI | 6 Aug 2026 |
| GPT-5.4 OpenAI · Low effort | 84.4% | Independent | Epoch AI | 15 Jul 2026 |
| GPT-5.5 OpenAI · Low effort | 84.4% | Independent | Epoch AI | 7 Aug 2026 |
| o3 OpenAI · Medium effort | 84.4% | Independent | Epoch AI | 7 Aug 2026 |
| Qwen3.5-Flash | 84.4% | Independent | Epoch AI | 7 Aug 2026 |
| Qwen3.6-Flash | 84.4% | Independent | Epoch AI | 7 Aug 2026 |
| Grok 4 | 84.0% | Independent | Epoch AI | — |
| o3 OpenAI · High effort | 83.9% | Independent | Epoch AI | 16 Apr 2025 |
| GLM-4.7 | 83.3% | Independent | Epoch AI | 29 Jan 2026 |
| Kimi K2 Thinking | 83.1% | Independent | Epoch AI | 13 Nov 2025 |
| Claude Sonnet 4.6 Anthropic · Medium effort | 82.2% | Independent | Epoch AI | 13 Jul 2026 |
| Gemini 3.6 Flash Google DeepMind · Low effort | 82.2% | Independent | Epoch AI | 7 Aug 2026 |
| Gemma 4 26B A4B Google DeepMind · Minimal effort | 82.2% | Independent | Epoch AI | 28 Aug 2026 |
| Qwen3.5-397B-A17B Alibaba Group · None effort | 82.2% | Independent | Epoch AI | 7 Aug 2026 |
| Claude Opus 4.5 Anthropic · 16K effort | 81.7% | Independent | Epoch AI | 24 Nov 2025 |
| o4-mini OpenAI · High effort | 81.7% | Independent | Epoch AI | 16 Apr 2025 |
| GPT-5 nano OpenAI · High effort | 81.1% | Independent | Epoch AI | 31 Oct 2025 |
| Claude Sonnet 5 Anthropic · Max effort | 80.0% | Independent | Epoch AI | 6 Aug 2026 |
| Gemini 3.1 Flash-Lite Google DeepMind · High effort | 80.0% | Independent | Epoch AI | 6 Aug 2026 |
| Gemini 3.5 Flash Google DeepMind · Minimal effort | 80.0% | Independent | Epoch AI | 15 Jul 2026 |
| Gemini 3.6 Flash Google DeepMind · Minimal effort | 80.0% | Independent | Epoch AI | 7 Aug 2026 |
| GLM-5 | 80.0% | Independent | Epoch AI | 12 Feb 2026 |
| Qwen3.7-Plus Alibaba Group · None effort | 80.0% | Independent | Epoch AI | 7 Aug 2026 |
| GPT-5.2 OpenAI · Low effort | 78.9% | Independent | Epoch AI | 11 Dec 2025 |
| GPT-5 mini OpenAI · Medium effort | 78.3% | Independent | Epoch AI | 7 Aug 2025 |
| Claude Sonnet 4.5 Anthropic · 59K effort | 77.8% | Independent | Epoch AI | 28 Oct 2025 |
| Claude Sonnet 4.5 Anthropic · 32K effort | 77.8% | Independent | Epoch AI | 21 Oct 2025 |
| Grok 3 Mini SpaceXAI (xAI) · High effort | 77.8% | Independent | Epoch AI | 10 Apr 2025 |
| Qwen3.7-Flash Alibaba Group · None effort | 77.8% | Independent | Epoch AI | 7 Aug 2026 |
| o3-mini OpenAI · High effort | 76.9% | Independent | Epoch AI | 27 Feb 2025 |
| Claude Sonnet 4.6 Anthropic · High effort | 75.6% | Independent | Epoch AI | 13 Jul 2026 |
| GLM-5.2 Z.ai (Zhipu AI) · Low effort | 75.6% | Independent | Epoch AI | 10 Aug 2026 |
| GPT-5 nano OpenAI · Medium effort | 74.2% | Independent | Epoch AI | 7 Aug 2025 |
| Gemma 4 31B Google DeepMind · Minimal effort | 73.3% | Independent | Epoch AI | 6 Aug 2026 |
| o1 OpenAI · High effort | 73.3% | Independent | Epoch AI | 15 Jul 2026 |
| o1 OpenAI · Medium effort | 73.3% | Independent | Epoch AI | 27 Feb 2025 |
| o4-mini OpenAI · Medium effort | 73.3% | Independent | Epoch AI | 7 Aug 2026 |
| Qwen3-Max | 73.3% | Independent | Epoch AI | 6 Oct 2025 |
| Gemini 2.5 Flash | 73.1% | Independent | Epoch AI | 22 Apr 2025 |
| Claude Sonnet 4 Anthropic · 32K effort | 71.1% | Independent | Epoch AI | 22 May 2025 |
| Claude Sonnet 4.5 Anthropic · 16K effort | 71.1% | Independent | Epoch AI | 28 Oct 2025 |
| Claude Sonnet 4.6 Anthropic · Max effort | 71.1% | Independent | Epoch AI | 6 Aug 2026 |
| Gemini 3.5 Flash-Lite Google DeepMind · High effort | 71.1% | Independent | Epoch AI | 6 Aug 2026 |
| MiniMax-M3 | 71.1% | Independent | Epoch AI | 10 Aug 2026 |
| Gemini 2.5 Flash | 70.8% | Independent | Epoch AI | 20 May 2025 |
| Qwen3-30B-A3B-Thinking-2507 | 70.3% | Independent | Epoch AI | 30 Aug 2026 |
| Qwen3.5-35B-A3B Alibaba Group · None effort | 70.0% | Independent | Epoch AI | 30 Aug 2026 |
| Claude Opus 4.1 Anthropic · 27K effort | 68.9% | Independent | Epoch AI | 5 Aug 2025 |
| Claude Sonnet 4 Anthropic · 59K effort | 68.9% | Independent | Epoch AI | 23 May 2025 |
| GPT-5.4 nano OpenAI · Low effort | 68.9% | Independent | Epoch AI | 7 Aug 2026 |
| GPT-5.6 Sol OpenAI · None effort | 68.9% | Independent | Epoch AI | 7 Aug 2026 |
| Kimi K3 Moonshot AI · Low effort | 68.9% | Independent | Epoch AI | 7 Aug 2026 |
| Qwen3.6-35B-A3B Alibaba Group · None effort | 68.9% | Independent | Epoch AI | 7 Aug 2026 |
| Seed-OSS-36B-Instruct | 67.5% | Independent | Epoch AI | 28 Aug 2026 |
| Qwen3-32B | 66.9% | Independent | Epoch AI | 30 Aug 2026 |
| Claude Haiku 4.5 Anthropic · 32K effort | 66.7% | Independent | Epoch AI | 22 Oct 2025 |
| GPT-5.6 Luna OpenAI · Low effort | 66.7% | Independent | Epoch AI | 7 Aug 2026 |
| Qwen3.6-27B Alibaba Group · None effort | 66.7% | Independent | Epoch AI | 7 Aug 2026 |
| DeepSeek-R1 | 66.4% | Independent | Epoch AI | 29 May 2025 |
| Qwen3-14B | 66.4% | Independent | Epoch AI | 28 Aug 2026 |
| gpt-oss-20b OpenAI · Medium effort | 65.3% | Independent | Epoch AI | 27 Aug 2026 |
| Claude Opus 4 Anthropic · 27K effort | 64.4% | Independent | Epoch AI | 28 May 2025 |
| Claude Opus 4.1 Anthropic · 16K effort | 64.4% | Independent | Epoch AI | 5 Aug 2025 |
| GPT-5.1 OpenAI · Low effort | 63.9% | Independent | Epoch AI | 25 Nov 2025 |
| o3-mini OpenAI · Medium effort | 63.9% | Independent | Epoch AI | 25 Feb 2025 |
| Qwen3-30B-A3B | 62.8% | Independent | Epoch AI | 28 Aug 2026 |
| GPT-5.2 OpenAI · None effort | 62.2% | Independent | Epoch AI | 13 Jul 2026 |
| Grok 3 Mini SpaceXAI (xAI) · Low effort | 62.2% | Independent | Epoch AI | 10 Apr 2025 |
| Qwen3-30B-A3B-Instruct-2507 | 62.2% | Independent | Epoch AI | 30 Aug 2026 |
| Qwen3.5-9B Alibaba Group · None effort | 61.7% | Independent | Epoch AI | 27 Aug 2026 |
| Claude Opus 4 Anthropic · 16K effort | 60.0% | Independent | Epoch AI | 22 May 2025 |
| Gemini 3.5 Flash-Lite Google DeepMind · Low effort | 60.0% | Independent | Epoch AI | 6 Aug 2026 |
| o3 OpenAI · Low effort | 60.0% | Independent | Epoch AI | 15 Jul 2026 |
| QwQ-32B | 59.2% | Independent | Epoch AI | 28 Aug 2026 |
| GLM-4.7-Flash | 58.3% | Independent | Epoch AI | 30 Aug 2026 |
| Claude 3.7 Sonnet Anthropic · 64K effort | 57.8% | Independent | Epoch AI | 13 Mar 2025 |
| Gemini 2.0 Flash Thinking | 57.8% | Independent | Epoch AI | 7 Mar 2025 |
| GPT-5.4 OpenAI · None effort | 57.8% | Independent | Epoch AI | 15 Jul 2026 |
| GPT-5.5 OpenAI · None effort | 57.8% | Independent | Epoch AI | 7 Aug 2026 |
| o4-mini OpenAI · Low effort | 57.8% | Independent | Epoch AI | 13 Jul 2026 |
| Qwen3-8B | 56.1% | Independent | Epoch AI | 27 Aug 2026 |
| Qwen3.5-4B Alibaba Group · None effort | 55.8% | Independent | Epoch AI | 28 Aug 2026 |
| DeepSeek-R1-Distill-Qwen-32B | 55.6% | Independent | Epoch AI | 28 Aug 2026 |
| GPT-5 mini OpenAI · Minimal effort | 55.6% | Independent | Epoch AI | 7 Aug 2026 |
| Grok 3 | 55.6% | Independent | Epoch AI | 10 Apr 2025 |
| Qwen3.5-35B-A3B | 54.4% | Independent | Epoch AI | 30 Aug 2026 |
| Claude 3.7 Sonnet Anthropic · 32K effort | 53.3% | Independent | Epoch AI | 12 Mar 2025 |
| Claude Sonnet 4 Anthropic · 16K effort | 53.3% | Independent | Epoch AI | 22 May 2025 |
| DeepSeek-R1 | 53.3% | Independent | Epoch AI | 26 Feb 2025 |
| GPT-5.6 Terra OpenAI · None effort | 53.3% | Independent | Epoch AI | 7 Aug 2026 |
| o1 OpenAI · Low effort | 53.3% | Independent | Epoch AI | 15 Jul 2026 |
| Qwen3-4B-Instruct-2507 | 52.2% | Independent | Epoch AI | 28 Aug 2026 |
| DeepSeek-R1-Distill-Llama-70B | 51.4% | Independent | Epoch AI | 7 Mar 2025 |
| Gemini 3.5 Flash-Lite Google DeepMind · Minimal effort | 51.1% | Independent | Epoch AI | 6 Aug 2026 |
| gpt-oss-20b OpenAI · High effort | 50.8% | Independent | Epoch AI | 27 Aug 2026 |
| DeepSeek-R1-Distill-Qwen-14B | 50.6% | Independent | Epoch AI | 28 Aug 2026 |
| DeepSeek-V3.2 | 48.9% | Independent | Epoch AI | 16 Jul 2026 |
| Claude Opus 4.5 | 48.1% | Independent | Epoch AI | 24 Nov 2025 |
| o1-mini OpenAI · High effort | 46.9% | Independent | Epoch AI | 6 Mar 2025 |
| Claude 3.7 Sonnet Anthropic · 16K effort | 46.7% | Independent | Epoch AI | 26 Feb 2025 |
| DeepSeek-V4-Pro DeepSeek · None effort | 46.7% | Independent | Epoch AI | 6 Aug 2026 |
| GPT-5 OpenAI · Minimal effort | 46.7% | Independent | Epoch AI | 20 Jul 2026 |
| GPT-5 nano OpenAI · Low effort | 46.7% | Independent | Epoch AI | 13 Jul 2026 |
| GPT-5.4 nano OpenAI · None effort | 46.7% | Independent | Epoch AI | 7 Aug 2026 |
| GPT-4.1 mini | 44.7% | Independent | Epoch AI | 14 Apr 2025 |
| o1-mini OpenAI · Medium effort | 44.7% | Independent | Epoch AI | 6 Mar 2025 |
| Gemini 3.1 Flash-Lite Google DeepMind · Low effort | 44.4% | Independent | Epoch AI | 6 Aug 2026 |
| o3-mini OpenAI · Low effort | 44.4% | Independent | Epoch AI | 20 Jul 2026 |
| DeepSeek-R1-0528-Qwen3-8B | 43.9% | Independent | Epoch AI | 28 Aug 2026 |
| Qwen3.5-9B | 42.8% | Independent | Epoch AI | 27 Aug 2026 |
| Claude Opus 4 | 42.2% | Independent | Epoch AI | 22 May 2025 |
| gpt-oss-20b OpenAI · Low effort | 40.3% | Independent | Epoch AI | 27 Aug 2026 |
| Claude Opus 4.1 | 40.0% | Independent | Epoch AI | 5 Aug 2025 |
| GPT-5.6 Luna OpenAI · None effort | 40.0% | Independent | Epoch AI | 7 Aug 2026 |
| GPT-4.1 | 38.3% | Independent | Epoch AI | 14 Apr 2025 |
| Gemini 3.1 Flash-Lite Google DeepMind · Minimal effort | 37.8% | Independent | Epoch AI | 6 Aug 2026 |
| GPT-4.5 | 37.8% | Independent | Epoch AI | 28 Feb 2025 |
| GPT-5.1 OpenAI · None effort | 37.8% | Independent | Epoch AI | 7 Aug 2026 |
| Claude Haiku 4.5 | 35.8% | Independent | Epoch AI | 16 Oct 2025 |
| Claude Sonnet 4.5 | 35.6% | Independent | Epoch AI | 29 Sep 2025 |
| GPT-5 nano OpenAI · Minimal effort | 35.6% | Independent | Epoch AI | 13 Jul 2026 |
| Mistral Medium 3 | 32.2% | Independent | Epoch AI | 7 May 2025 |
| Gemini 2.0 Flash | 31.1% | Independent | Epoch AI | 25 Feb 2025 |
| o1-preview | 31.1% | Independent | Epoch AI | 7 Mar 2025 |
| Magistral Small 1.0 | 30.0% | Independent | Epoch AI | 6 Aug 2026 |
| Claude Sonnet 4 | 28.9% | Independent | Epoch AI | 22 May 2025 |
| GLM-5.2 Z.ai (Zhipu AI) · None effort | 28.9% | Independent | Epoch AI | 10 Aug 2026 |
| GPT-4.1 nano | 28.9% | Independent | Epoch AI | 14 Apr 2025 |
| Magistral Small 1.2 | 28.1% | Independent | Epoch AI | 30 Aug 2026 |
| GPT-5.4 mini OpenAI · None effort | 26.7% | Independent | Epoch AI | 7 Aug 2026 |
| MiniMax-M3 MiniMax · None effort | 26.7% | Independent | Epoch AI | 10 Aug 2026 |
| Qwen3-14B Alibaba Group · None effort | 25.8% | Independent | Epoch AI | 30 Aug 2026 |
| Qwen3-30B-A3B Alibaba Group · None effort | 25.6% | Independent | Epoch AI | 30 Aug 2026 |
| GLM-4.7-Flash Z.ai (Zhipu AI) · None effort | 25.0% | Independent | Epoch AI | 28 Aug 2026 |
| Gemini 1.5 Pro | 23.1% | Independent | Epoch AI | 25 Feb 2025 |
| Qwen3-32B Alibaba Group · None effort | 23.1% | Independent | Epoch AI | 30 Aug 2026 |
| Gemma 3 27B | 22.5% | Independent | Epoch AI | 27 Aug 2026 |
| Qwen3-8B Alibaba Group · None effort | 22.2% | Independent | Epoch AI | 27 Aug 2026 |
| Claude 3.7 Sonnet | 21.9% | Independent | Epoch AI | 25 Feb 2025 |
| DeepSeek-R1-Distill-Qwen-1.5B | 21.4% | Independent | Epoch AI | 28 Aug 2026 |
| Llama 4 Maverick | 20.6% | Independent | Epoch AI | 8 Apr 2025 |
| Qwen-Plus | 17.8% | Independent | Epoch AI | 7 Apr 2025 |
| Gemma 3 12B | 16.7% | Independent | Epoch AI | 28 Aug 2026 |
| Gemini 1.5 Flash | 16.3% | Independent | Epoch AI | 25 Feb 2025 |
| Qwen2.5-Max | 16.1% | Independent | Epoch AI | 1 Apr 2025 |
| DeepSeek-V3 | 15.8% | Independent | Epoch AI | 25 Feb 2025 |
| Phi-4 | 13.8% | Independent | Epoch AI | 25 Feb 2025 |
| Grok 2 | 11.5% | Independent | Epoch AI | 25 Feb 2025 |
| Llama 3.1 405B | 9.7% | Independent | Epoch AI | 25 Feb 2025 |
| Claude 3.5 Sonnet | 8.5% | Independent | Epoch AI | 25 Feb 2025 |
| Mistral Large 2.0 | 8.5% | Independent | Epoch AI | 25 Feb 2025 |
| Qwen2.5-72B-Instruct | 8.1% | Independent | Epoch AI | 25 Feb 2025 |
| Qwen3-1.7B Alibaba Group · None effort | 8.1% | Independent | Epoch AI | 28 Aug 2026 |
| Llama 4 Scout | 7.8% | Independent | Epoch AI | 8 Apr 2025 |
| Mistral Large 2.1 | 7.8% | Independent | Epoch AI | 25 Feb 2025 |
| Gemma 3 4B | 7.5% | Independent | Epoch AI | 28 Aug 2026 |
| Qwen2.5-32B-Instruct | 7.4% | Independent | Epoch AI | 25 Feb 2025 |
| GPT-4o mini | 6.9% | Independent | Epoch AI | 30 Jul 2025 |
| Gemini 1.5 Pro | 6.8% | Independent | Epoch AI | 25 Feb 2025 |
| GPT-4 Turbo | 6.7% | Independent | Epoch AI | 27 Feb 2025 |
| Claude 3.5 Sonnet | 6.5% | Independent | Epoch AI | 25 Feb 2025 |
| GPT-4o | 6.4% | Independent | Epoch AI | 25 Feb 2025 |
| GPT-4o | 6.3% | Independent | Epoch AI | 25 Feb 2025 |
| GPT-4o | 6.3% | Independent | Epoch AI | 25 Feb 2025 |
| Qwen-Turbo | 6.1% | Independent | Epoch AI | 7 Apr 2025 |
| Mistral Small 3.1 | 5.8% | Independent | Epoch AI | 18 Mar 2025 |
| Mistral Small 3.0 | 5.3% | Independent | Epoch AI | 18 Mar 2025 |
| Llama 3.3 70B | 5.1% | Independent | Epoch AI | 25 Feb 2025 |
| Claude 3 Opus | 4.7% | Independent | Epoch AI | 25 Feb 2025 |
| Gemini 1.5 Flash-8B | 4.6% | Independent | Epoch AI | 25 Feb 2025 |
| Claude 3.5 Haiku | 4.3% | Independent | Epoch AI | 25 Feb 2025 |
| Llama 3 70B | 4.3% | Independent | Epoch AI | 25 Feb 2025 |
| Gemini 1.5 Flash | 3.9% | Independent | Epoch AI | 25 Feb 2025 |
| Llama 3.1 70B | 3.6% | Independent | Epoch AI | 25 Feb 2025 |
| Granite 4.0 Micro (3B) | 2.8% | Independent | Epoch AI | 30 Aug 2026 |
| Llama 3.2 90B Vision | 2.6% | Independent | Epoch AI | 25 Feb 2025 |
| Claude 2 | 2.5% | Independent | Epoch AI | 7 Mar 2025 |
| Claude 3 Sonnet | 2.5% | Independent | Epoch AI | 25 Feb 2025 |
| Hermes 2 Theta Llama 3 70B | 2.5% | Independent | Epoch AI | 1 Apr 2025 |
| Qwen2.5-7B-Instruct | 2.5% | Independent | Epoch AI | 30 Aug 2026 |
| GPT-3.5 Turbo | 2.2% | Independent | Epoch AI | 7 Aug 2026 |
| Claude 2.1 | 1.9% | Independent | Epoch AI | 7 Mar 2025 |
| Llama 3 8B | 1.9% | Independent | Epoch AI | 30 Aug 2026 |
| Mistral Large 1.0 | 1.9% | Independent | Epoch AI | 25 Feb 2025 |
| Claude 3 Haiku | 1.8% | Independent | Epoch AI | 25 Feb 2025 |
| Llama 3.1 8B | 1.7% | Independent | Epoch AI | 27 Aug 2026 |
| Gemma 2 27B | 1.4% | Independent | Epoch AI | 7 Mar 2025 |
| Gemini 1.0 Pro | 1.1% | Independent | Epoch AI | 25 Feb 2025 |
| Gemma 3 1B | 1.1% | Independent | Epoch AI | 28 Aug 2026 |
| GPT-4 | 1.1% | Independent | Epoch AI | 23 Oct 2025 |
| DeepSeek LLM 67B Chat | 0.8% | Independent | Epoch AI | 28 Aug 2026 |
| Granite 4.0 1B (Nano) | 0.8% | Independent | Epoch AI | 30 Aug 2026 |
| Gemma 2 9B | 0.6% | Independent | Epoch AI | 7 Mar 2025 |
| GPT-4 | 0.6% | Independent | Epoch AI | 22 Oct 2025 |
| Llama 3.2 1B | 0.6% | Independent | Epoch AI | 30 Aug 2026 |
| Granite 4.0 350M (Nano) | 0.3% | Independent | Epoch AI | 28 Aug 2026 |
| Mistral 7B v0.3 | 0.3% | Independent | Epoch AI | 30 Aug 2026 |
| Llama 2 70B | 0.0% | Independent | Epoch AI | 25 Feb 2025 |
- Epoch AI, 'Capabilities & benchmarking', epoch.ai/benchmarks, CC BY 4.0.