Chess Puzzles
Re-run by Epoch AINo description yet: we only describe an eval set when we can do it correctly.
- Kind
- Run independently by Epoch AI, with public logs. Our own re-runs are not built yet.
- Score shown
- Epoch's column "Best score (across scorers)", as a percentage, higher is better; random guessing scores about 5.0%. Error bars as the source gives them.
- Results
- 166 in all: 25 for open-weight models, 138 for closed ones, 3 where Epoch does not say. 23 linked to a model page here.
Best open-weight result: deepseek-ai/DeepSeek-V4.1-Flash, 48.0%, number 24 of 166 results.
| # | Model | Weights | Setting | Score | Date | Source |
|---|---|---|---|---|---|---|
| 1 | claude-fable-5-2_maxAnthropic | closed | effort max | 79.0%±4.1 | 6 Nov 2026 | Epoch AI |
| 2 | gemini-4-carbon_highGoogle DeepMind | closed | effort high | 78.0%±4.2 | 2 Nov 2026 | Epoch AI |
| 3 | gpt-6.2-astra_maxOpenAI | closed | effort max | 78.0%±4.2 | 27 Oct 2026 | Epoch AI |
| 4 | gpt-6-astra_promaxOpenAI | closed | default | 76.0%±4.3 | 15 Oct 2026 | Epoch AI log |
| 5 | gpt-6-astra_maxOpenAI | closed | effort max | 74.0%±4.4 | 13 Oct 2026 | Epoch AI log |
| 6 | gpt-5.6-sol_promaxOpenAI | closed | default | 68.0%±4.7 | 14 Oct 2026 | Epoch AI log |
| 7 | gpt-6.1-sol_maxOpenAI | closed | effort max | 68.0%±4.7 | 13 Oct 2026 | Epoch AI log |
| 8 | claude-opus-5-5_maxAnthropic | closed | effort max | 67.0%±4.7 | 13 Oct 2026 | Epoch AI log |
| 9 | gemini-3.7-flash_highGoogle DeepMind | closed | effort high | 66.0%±4.8 | 13 Oct 2026 | Epoch AI log |
| 10 | gpt-5.5_xhighOpenAI | closed | effort xhigh | 63.0%±4.9 | 13 Oct 2026 | Epoch AI log |
| 11 | gpt-5.5-pro_xhighOpenAI | closed | effort xhigh | 62.0%±4.9 | 14 Oct 2026 | Epoch AI log |
| 12 | gemini-4-hydrogen_xhighGoogle DeepMind | closed | effort xhigh | 60.0%±4.9 | 13 Nov 2026 | Epoch AI log |
| 13 | gpt-5.2-pro-2025-12-11_xhighOpenAI | closed | effort xhigh | 60.0%±4.9 | 14 Oct 2026 | Epoch AI log |
| 14 | claude-sonnet-5-5_maxAnthropic | closed | effort max | 59.0%±4.9 | 13 Oct 2026 | Epoch AI log |
| 15 | gemini-4-argon_highGoogle DeepMind | closed | effort high | 59.0%±4.9 | 13 Oct 2026 | Epoch AI log |
| 16 | gpt-5.4-pro-2026-03-05_xhighOpenAI | closed | effort xhigh | 59.0%±4.9 | 14 Oct 2026 | Epoch AI log |
| 17 | gpt-5.6-sol_maxOpenAI | closed | effort max | 59.0%±4.9 | 13 Oct 2026 | Epoch AI log |
| 18 | gemini-3.8-flash_highGoogle DeepMind | closed | effort high | 58.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 19 | gpt-6-sol_maxOpenAI | closed | effort max | 56.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 20 | claude-haiku-5-5_maxAnthropic | closed | effort max | 55.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 21 | claude-opus-4-8_maxAnthropic | closed | effort max | 52.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 22 | gemini-3.1-pro-preview_highGoogle DeepMind | closed | effort high | 51.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 23 | claude-fable-5-1_maxAnthropic | closed | effort max | 48.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 24 | deepseek-ai/DeepSeek-V4.1-FlashEpoch: deepseek-v4.1-flash_max, DeepSeek | open | effort max | 48.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 25 | gpt-5-pro-2025-10-06_highOpenAI | closed | effort high | 48.0%±5.0 | 14 Oct 2026 | Epoch AI log |
| 26 | gpt-5.2-2025-12-11_xhighOpenAI | closed | effort xhigh | 48.0%±5.0 | 14 Oct 2026 | Epoch AI log |
| 27 | gpt-5.4-2026-03-05_xhighOpenAI | closed | effort xhigh | 48.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 28 | claude-fable-5_maxAnthropic | closed | effort max | 47.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 29 | gpt-5.6-terra_maxOpenAI | closed | effort max | 47.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 30 | deepseek-ai/DeepSeek-V4-Pro-0813Epoch: deepseek-v4-pro-0813_max, DeepSeek | open | effort max | 46.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 31 | deepseek-ai/DeepSeek-V4.1-Flash-1017Epoch: deepseek-v4.1-flash-1017_max, DeepSeek | open | effort max | 46.0%±5.0 | 19 Oct 2026 | Epoch AI log |
| 32 | qwen3.8-max_xhighAlibaba | closed | effort xhigh | 46.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 33 | claude-fable-5_highAnthropic | closed | effort high | 45.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 34 | claude-sonnet-5_maxAnthropic | closed | effort max | 45.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 35 | gemini-3.5-flash_highGoogle DeepMind | closed | effort high | 45.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 36 | gpt-5.2-2025-12-11_highOpenAI | closed | effort high | 45.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 37 | gpt-5.4-nano-2026-03-17_xhighOpenAI | closed | effort xhigh | 45.0%±5.0 | 14 Oct 2026 | Epoch AI log |
| 38 | moonshotai/Kimi-K3.1-ProEpoch: kimi-k3.1-pro-together_max, Moonshot | open | effort max | 45.0%±5.0 | 6 Nov 2026 | Epoch AI log |
| 39 | claude-opus-5_maxAnthropic | closed | effort max | 43.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 40 | gpt-5.6-luna_maxOpenAI | closed | effort max | 43.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 41 | gemini-3.5-flash_lowGoogle DeepMind | closed | effort low | 42.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 42 | gemini-3.6-flash_lowGoogle DeepMind | closed | effort low | 42.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 43 | gpt-5.1-2025-11-13_highOpenAI | closed | effort high | 42.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 44 | gpt-5.4-2026-03-05_highOpenAI | closed | effort high | 42.0%±5.0 | 13 Oct 2026 | Epoch AI log |
| 45 | claude-opus-4-7_maxAnthropic | closed | effort max | 41.0%±4.9 | 13 Oct 2026 | Epoch AI log |
| 46 | grok-4.7_xhighxAI | closed | effort xhigh | 41.0%±4.9 | 13 Oct 2026 | Epoch AI log |
| 47 | doubao-seed-evolving_highByteDance | closed | effort high | 40.0%±4.9 | 2 Nov 2026 | Epoch AI log |
| 48 | gemini-3-flash-preview_highGoogle DeepMind | closed | effort high | 40.0%±4.9 | 13 Oct 2026 | Epoch AI log |
| 49 | gemini-3.6-flash_highGoogle DeepMind | closed | effort high | 40.0%±4.9 | 13 Oct 2026 | Epoch AI log |
| 50 | zai-org/GLM-5.3Epoch: glm-5.3_max, Z.ai (Zhipu AI) | open | effort max | 37.0%±4.9 | 13 Oct 2026 | Epoch AI log |
| 51 | gpt-5.4-2026-03-05_mediumOpenAI | closed | effort medium | 37.0%±4.9 | 13 Oct 2026 | Epoch AI log |
| 52 | moonshotai/Kimi-K3Epoch: kimi-k3_high, Moonshot | open | effort high | 37.0%±4.9 | 13 Oct 2026 | Epoch AI log |
| 53 | muse-spark-1.3_maxMeta AI | closed | effort max | 37.0%±4.9 | 13 Oct 2026 | Epoch AI log |
| 54 | o3-2025-04-16_mediumOpenAI | closed | effort medium | 37.0%±4.9 | 13 Oct 2026 | Epoch AI log |
| 55 | qwen3.8-max-0902_xhighAlibaba | closed | effort xhigh | 37.0%±4.9 | 13 Oct 2026 | Epoch AI log |
| 56 | claude-opus-4-7_xhighAnthropic | closed | effort xhigh | 36.0%±4.8 | 13 Oct 2026 | Epoch AI log |
| 57 | gpt-5-2025-08-07_highOpenAI | closed | effort high | 36.0%±4.8 | 13 Oct 2026 | Epoch AI log |
| 58 | gpt-5.2-2025-12-11_mediumOpenAI | closed | effort medium | 36.0%±4.8 | 13 Oct 2026 | Epoch AI log |
| 59 | gpt-5.4-nano-2026-03-17_highOpenAI | closed | effort high | 36.0%±4.8 | 13 Oct 2026 | Epoch AI log |
| 60 | gpt-6-luna_maxOpenAI | closed | effort max | 36.0%±4.8 | 13 Oct 2026 | Epoch AI log |
| 61 | moonshotai/Kimi-K3Epoch: kimi-k3_max, Moonshot | open | effort max | 36.0%±4.8 | 13 Oct 2026 | Epoch AI log |
| 62 | o3-2025-04-16_highOpenAI | closed | effort high | 36.0%±4.8 | 13 Oct 2026 | Epoch AI log |
| 63 | o3-2025-04-16_lowOpenAI | closed | effort low | 36.0%±4.8 | 13 Oct 2026 | Epoch AI log |
| 64 | grok-4.5_highxAI | closed | effort high | 35.0%±4.8 | 13 Oct 2026 | Epoch AI log |
| 65 | claude-sonnet-5_xhighAnthropic | closed | effort xhigh | 34.0%±4.8 | 13 Oct 2026 | Epoch AI log |
| 66 | muse-spark-1.1_xhighMeta AI | closed | effort xhigh | 34.0%±4.8 | 13 Oct 2026 | Epoch AI log |
| 67 | muse-spark-1.3_xhighMeta AI | closed | effort xhigh | 34.0%±4.8 | 13 Oct 2026 | Epoch AI log |
| 68 | grok-4.6_highxAI | closed | effort high | 33.0%±4.7 | 13 Oct 2026 | Epoch AI log |
| 69 | grok-4.6_xhighxAI | closed | effort xhigh | 33.0%±4.7 | 13 Oct 2026 | Epoch AI log |
| 70 | mistral-large-4-0_highMistral AIno page match | open | effort high | 33.0%±4.7 | 15 Oct 2026 | Epoch AI log |
| 71 | deepseek-ai/DeepSeek-V4-Flash-0731Epoch: deepseek-v4-flash-0731_max, DeepSeek | open | effort max | 32.0%±4.7 | 13 Oct 2026 | Epoch AI log |
| 72 | gpt-5-2025-08-07_mediumOpenAI | closed | effort medium | 32.0%±4.7 | 13 Oct 2026 | Epoch AI log |
| 73 | claude-fable-5_lowAnthropic | closed | effort low | 31.0%±4.6 | 13 Oct 2026 | Epoch AI log |
| 74 | gpt-5.6-sol_lowOpenAI | closed | effort low | 31.0%±4.6 | 13 Oct 2026 | Epoch AI log |
| 75 | gpt-5.5_lowOpenAI | closed | effort low | 30.0%±4.6 | 13 Oct 2026 | Epoch AI log |
| 76 | claude-opus-4-8_lowAnthropic | closed | effort low | 29.0%±4.6 | 13 Oct 2026 | Epoch AI log |
| 77 | gemini-3.6-flash_minimalGoogle DeepMind | closed | effort minimal | 29.0%±4.6 | 13 Oct 2026 | Epoch AI log |
| 78 | grok-4.20-0309-reasoningxAI | closed | default | 29.0%±4.6 | 13 Oct 2026 | Epoch AI log |
| 79 | mistral-large-4-0-oss_highMistral AIno page match | open | effort high | 29.0%±4.6 | 2 Nov 2026 | Epoch AI log |
| 80 | claude-opus-5_lowAnthropic | closed | effort low | 27.0%±4.5 | 13 Oct 2026 | Epoch AI log |
| 81 | gemini-3.5-flash_minimalGoogle DeepMind | closed | effort minimal | 27.0%±4.5 | 13 Oct 2026 | Epoch AI log |
| 82 | gpt-5-2025-08-07_lowOpenAI | closed | effort low | 27.0%±4.5 | 13 Oct 2026 | Epoch AI log |
| 83 | gpt-5.6-terra_lowOpenAI | closed | effort low | 27.0%±4.5 | 13 Oct 2026 | Epoch AI log |
| 84 | muse-spark-1.2_xhighMeta AI | closed | effort xhigh | 27.0%±4.5 | 13 Oct 2026 | Epoch AI log |
| 85 | qwen3.7-maxAlibaba | closed | default | 27.0%±4.5 | 13 Oct 2026 | Epoch AI log |
| 86 | claude-opus-4-7_lowAnthropic | closed | effort low | 26.0%±4.4 | 13 Oct 2026 | Epoch AI log |
| 87 | doubao-seed-2.1-turbo-260628_highByteDance | closed | effort high | 26.0%±4.4 | 30 Oct 2026 | Epoch AI log |
| 88 | gpt-5.2-2025-12-11_lowOpenAI | closed | effort low | 26.0%±4.4 | 13 Oct 2026 | Epoch AI log |
| 89 | claude-haiku-5-5_lowAnthropic | closed | effort low | 24.0%±4.3 | 13 Oct 2026 | Epoch AI log |
| 90 | gpt-5-mini-2025-08-07_highOpenAI | closed | effort high | 24.0%±4.3 | 13 Oct 2026 | Epoch AI log |
| 91 | claude-sonnet-4-6_maxAnthropic | closed | effort max | 23.0%±4.2 | 13 Oct 2026 | Epoch AI log |
| 92 | gpt-5.1-2025-11-13_lowOpenAI | closed | effort low | 23.0%±4.2 | 13 Oct 2026 | Epoch AI log |
| 93 | gpt-5.4-mini-2026-03-17_highOpenAI | closed | effort high | 23.0%±4.2 | 13 Oct 2026 | Epoch AI log |
| 94 | zai-org/GLM-5.2Epoch: glm-5.2_max, Z.ai (Zhipu AI) | open | effort max | 22.0%±4.2 | 14 Oct 2026 | Epoch AI log |
| 95 | gpt-5-nano-2025-08-07_highOpenAI | closed | effort high | 22.0%±4.2 | 13 Oct 2026 | Epoch AI log |
| 96 | o4-mini-2025-04-16_mediumOpenAI | closed | effort medium | 22.0%±4.2 | 13 Oct 2026 | Epoch AI log |
| 97 | claude-opus-4-6_maxAnthropic | closed | effort max | 21.0%±4.1 | 13 Oct 2026 | Epoch AI log |
| 98 | doubao-seed-2.1-pro-260628_highByteDance | closed | effort high | 21.0%±4.1 | 28 Oct 2026 | Epoch AI log |
| 99 | gemini-3.5-flash-lite_lowGoogle DeepMind | closed | effort low | 21.0%±4.1 | 13 Oct 2026 | Epoch AI log |
| 100 | zai-org/GLM-4.7Epoch: glm-4.7, Z.ai (Zhipu AI) | open | default | 21.0%±4.1 | 14 Oct 2026 | Epoch AI log |
| 101 | zai-org/GLM-5.3-FlashEpoch: glm-5.3-flash_max, Z.ai (Zhipu AI) | open | effort max | 21.0%±4.1 | 13 Oct 2026 | Epoch AI log |
| 102 | thinkingmachines/InklingEpoch: Inkling_xhigh, Thinking Machines | open | effort xhigh | 21.0%±4.1 | 13 Oct 2026 | Epoch AI log |
| 103 | doubao-seed-2.0-lite-260428_highByteDance | closed | effort high | 20.0%±4.0 | 28 Oct 2026 | Epoch AI log |
| 104 | doubao-seed-2.0-pro-260215_highByteDance | closed | effort high | 20.0%±4.0 | 28 Oct 2026 | Epoch AI log |
| 105 | zai-org/GLM-5.1Epoch: glm-5.1, Z.ai (Zhipu AI) | open | default | 20.0%±4.0 | 14 Oct 2026 | Epoch AI log |
| 106 | gpt-5.4-nano-2026-03-17_lowOpenAI | closed | effort low | 20.0%±4.0 | 13 Oct 2026 | Epoch AI log |
| 107 | moonshotai/Kimi-K2.6Epoch: kimi-k2.6, Moonshot | open | default | 20.0%±4.0 | 13 Oct 2026 | Epoch AI log |
| 108 | claude-opus-4-8_noneAnthropic | closed | effort none | 19.0%±3.9 | 13 Oct 2026 | Epoch AI log |
| 109 | claude-sonnet-4-6_highAnthropic | closed | effort high | 19.0%±3.9 | 13 Oct 2026 | Epoch AI log |
| 110 | deepseek-ai/DeepSeek-V4-ProEpoch: deepseek-v4-pro_max, DeepSeek | open | effort max | 19.0%±3.9 | 13 Oct 2026 | Epoch AI log |
| 111 | zai-org/GLM-5.2Epoch: glm-5.2_none, Z.ai (Zhipu AI) | open | effort none | 19.0%±3.9 | 14 Oct 2026 | Epoch AI log |
| 112 | gpt-5.4-2026-03-05_lowOpenAI | closed | effort low | 19.0%±3.9 | 13 Oct 2026 | Epoch AI log |
| 113 | zai-org/GLM-5.2Epoch: glm-5.2_low, Z.ai (Zhipu AI) | open | effort low | 18.0%±3.9 | 14 Oct 2026 | Epoch AI log |
| 114 | grok-4.3_highxAI | closed | effort high | 18.0%±3.9 | 13 Oct 2026 | Epoch AI log |
| 115 | moonshotai/Kimi-K3Epoch: kimi-k3_low, Moonshot | open | effort low | 18.0%±3.9 | 13 Oct 2026 | Epoch AI log |
| 116 | gpt-5.4-mini-2026-03-17_xhighOpenAI | closed | effort xhigh | 17.0%±3.8 | 13 Oct 2026 | Epoch AI log |
| 117 | gpt-5.6-luna_lowOpenAI | closed | effort low | 17.0%±3.8 | 13 Oct 2026 | Epoch AI log |
| 118 | MiniMaxAI/MiniMax-M3Epoch: MiniMax-M3, MiniMax | open | default | 17.0%±3.8 | 13 Oct 2026 | Epoch AI log |
| 119 | claude-sonnet-4-6_mediumAnthropic | closed | effort medium | 16.0%±3.7 | 13 Oct 2026 | Epoch AI log |
| 120 | deepseek-ai/DeepSeek-V4-ProEpoch: deepseek-v4-pro_high, DeepSeek | open | effort high | 16.0%±3.7 | 13 Oct 2026 | Epoch AI log |
| 121 | gemini-3.5-flash-lite_highGoogle DeepMind | closed | effort high | 16.0%±3.7 | 13 Oct 2026 | Epoch AI log |
| 122 | gpt-5-mini-2025-08-07_lowOpenAI | closed | effort low | 16.0%±3.7 | 13 Oct 2026 | Epoch AI log |
| 123 | qwen3.6-plusAlibaba | closed | default | 16.0%±3.7 | 13 Oct 2026 | Epoch AI log |
| 124 | gpt-5-nano-2025-08-07_lowOpenAI | closed | effort low | 15.0%±3.6 | 13 Oct 2026 | Epoch AI log |
| 125 | o3-mini-2025-01-31_highOpenAI | closed | effort high | 15.0%±3.6 | 13 Oct 2026 | Epoch AI log |
| 126 | o4-mini-2025-04-16_lowOpenAI | closed | effort low | 15.0%±3.6 | 13 Oct 2026 | Epoch AI log |
| 127 | claude-opus-4-5-20251101_59KAnthropic | closed | default | 14.0%±3.5 | 13 Oct 2026 | Epoch AI log |
| 128 | moonshotai/Kimi-K2.7-CodeEpoch: kimi-k2.7-code, Moonshot | open | default | 14.0%±3.5 | 13 Oct 2026 | Epoch AI log |
| 129 | qwen3.5-flashAlibaba | not stated | default | 14.0%±3.5 | 13 Oct 2026 | Epoch AI log |
| 130 | o4-mini-2025-04-16_highOpenAI | closed | effort high | 13.0%±3.4 | 13 Oct 2026 | Epoch AI log |
| 131 | qwen3.6-flashAlibaba | not stated | default | 12.0%±3.3 | 13 Oct 2026 | Epoch AI log |
| 132 | gpt-4.1-2025-04-14OpenAI | closed | default | 11.0%±3.1 | 13 Oct 2026 | Epoch AI log |
| 133 | gpt-5-2025-08-07_minimalOpenAI | closed | effort minimal | 11.0%±3.1 | 13 Oct 2026 | Epoch AI log |
| 134 | o1-2024-12-17_lowOpenAI | closed | effort low | 11.0%±3.1 | 13 Oct 2026 | Epoch AI log |
| 135 | o3-mini-2025-01-31_mediumOpenAI | closed | effort medium | 11.0%±3.1 | 13 Oct 2026 | Epoch AI log |
| 136 | claude-sonnet-4-5-20250929_32KAnthropic | closed | default | 10.0%±3.0 | 13 Oct 2026 | Epoch AI log |
| 137 | claude-sonnet-4-5-20250929_59KAnthropic | closed | default | 10.0%±3.0 | 13 Oct 2026 | Epoch AI log |
| 138 | o1-2024-12-17_highOpenAI | closed | effort high | 10.0%±3.0 | 13 Oct 2026 | Epoch AI log |
| 139 | gpt-4o-2024-08-06OpenAI | closed | default | 9.0%±2.9 | 13 Oct 2026 | Epoch AI log |
| 140 | gpt-5.5_noneOpenAI | closed | effort none | 9.0%±2.9 | 13 Oct 2026 | Epoch AI log |
| 141 | qwen3.5-plusAlibaba | not stated | default | 9.0%±2.9 | 13 Oct 2026 | Epoch AI log |
| 142 | claude-haiku-4-5-20251001_59KAnthropic | closed | default | 8.0%±2.7 | 13 Oct 2026 | Epoch AI log |
| 143 | MiniMaxAI/MiniMax-M3Epoch: MiniMax-M3_none, MiniMax | open | effort none | 8.0%±2.7 | 13 Oct 2026 | Epoch AI log |
| 144 | o1-2024-12-17_mediumOpenAI | closed | effort medium | 8.0%±2.7 | 13 Oct 2026 | Epoch AI log |
| 145 | gemini-3.5-flash-lite_minimalGoogle DeepMind | closed | effort minimal | 7.0%±2.6 | 13 Oct 2026 | Epoch AI log |
| 146 | gpt-5.6-sol_noneOpenAI | closed | effort none | 7.0%±2.6 | 13 Oct 2026 | Epoch AI log |
| 147 | claude-sonnet-4-5-20250929Anthropic | closed | default | 6.0%±2.4 | 13 Oct 2026 | Epoch AI log |
| 148 | gpt-5.1-2025-11-13_noneOpenAI | closed | effort none | 6.0%±2.4 | 13 Oct 2026 | Epoch AI log |
| 149 | o3-mini-2025-01-31_lowOpenAI | closed | effort low | 6.0%±2.4 | 13 Oct 2026 | Epoch AI log |
| 150 | claude-haiku-4-5-20251001_32KAnthropic | closed | default | 5.0%±2.2 | 13 Oct 2026 | Epoch AI log |
| 151 | claude-opus-4-5-20251101Anthropic | closed | default | 5.0%±2.2 | 13 Oct 2026 | Epoch AI log |
| 152 | claude-opus-4-5-20251101_32KAnthropic | closed | default | 5.0%±2.2 | 13 Oct 2026 | Epoch AI log |
| 153 | gpt-5.6-terra_noneOpenAI | closed | effort none | 5.0%±2.2 | 13 Oct 2026 | Epoch AI log |
| 154 | seed-1.6ByteDance | closed | default | 5.0%±2.2 | 30 Oct 2026 | Epoch AI log |
| 155 | claude-3-opus-20240229Anthropic | closed | default | 4.0%±2.0 | 13 Oct 2026 | Epoch AI log |
| 156 | deepseek-ai/DeepSeek-V4-ProEpoch: deepseek-v4-pro_none, DeepSeek | open | effort none | 4.0%±2.0 | 13 Oct 2026 | Epoch AI log |
| 157 | gpt-4o-mini-2024-07-18OpenAI | closed | default | 3.0%±1.7 | 13 Oct 2026 | Epoch AI log |
| 158 | gpt-5.4-2026-03-05_noneOpenAI | closed | effort none | 3.0%±1.7 | 13 Oct 2026 | Epoch AI log |
| 159 | gpt-5.6-luna_noneOpenAI | closed | effort none | 3.0%±1.7 | 13 Oct 2026 | Epoch AI log |
| 160 | gpt-5-nano-2025-08-07_minimalOpenAI | closed | effort minimal | 2.0%±1.4 | 13 Oct 2026 | Epoch AI log |
| 161 | gpt-5.2-2025-12-11_noneOpenAI | closed | effort none | 2.0%±1.4 | 13 Oct 2026 | Epoch AI log |
| 162 | gpt-5.4-mini-2026-03-17_noneOpenAI | closed | effort none | 2.0%±1.4 | 13 Oct 2026 | Epoch AI log |
| 163 | gpt-5.4-nano-2026-03-17_noneOpenAI | closed | effort none | 2.0%±1.4 | 13 Oct 2026 | Epoch AI log |
| 164 | gpt-4.1-mini-2025-04-14OpenAI | closed | default | 1.0%±1.0 | 13 Oct 2026 | Epoch AI log |
| 165 | gpt-4.1-nano-2025-04-14OpenAI | closed | default | 0.0% | 13 Oct 2026 | Epoch AI log |
| 166 | gpt-5-mini-2025-08-07_minimalOpenAI | closed | effort minimal | 0.0% | 13 Oct 2026 | Epoch AI log |
Open-weight rows are shaded. Dates are when the result was run or evaluated, where the data says. Data from Epoch AI (CC-BY 4.0), fetched 19 Nov 2026: Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource]. Epoch AI. Method