CL-bench Life
ReportedNo description yet: we only describe an eval set when we can do it correctly.
- Kind
- Published by a third party's leaderboard, collected by Epoch AI. The leaderboard is named and linked.
- Score shown
- Epoch's column "Overall", as a percentage, higher is better; random guessing scores about 0.0%. The source gives no error bars.
- Results
- 17 in all: 6 for open-weight models, 10 for closed ones, 1 where Epoch does not say. 6 linked to a model page here.
Best open-weight result: deepseek-ai/DeepSeek-V4-Pro, 13.5%, number 9 of 17 results.
| # | Model | Weights | Setting | Score | Date | Source |
|---|---|---|---|---|---|---|
| 1 | gpt-5.5_highOpenAI | closed | effort high | 22.2% | not dated | via Epoch AI |
| 2 | gpt-5.4-2026-03-05_xhighOpenAI | closed | effort xhigh | 21.7% | not dated | via Epoch AI |
| 3 | gpt-5.4-2026-03-05_highOpenAI | closed | effort high | 19.3% | not dated | via Epoch AI |
| 4 | gpt-5.1-2025-11-13_highOpenAI | closed | effort high | 17.3% | not dated | via Epoch AI |
| 5 | claude-opus-4-6_highAnthropic | closed | effort high | 17.0% | not dated | via Epoch AI |
| 6 | gemini-3.1-pro-previewGoogle DeepMind | closed | default | 16.9% | not dated | via Epoch AI |
| 7 | gpt-5.4-2026-03-05_unknownOpenAI | closed | effort unknown | 13.8% | not dated | via Epoch AI |
| 8 | claude-opus-4-6_unknownAnthropic | closed | effort unknown | 13.6% | not dated | via Epoch AI |
| 9 | deepseek-ai/DeepSeek-V4-ProEpoch: deepseek-v4-pro_high, DeepSeek | open | effort high | 13.5% | not dated | via Epoch AI |
| 10 | moonshotai/Kimi-K2.5Epoch: kimi-k2.5, Moonshot | open | default | 13.2% | not dated | via Epoch AI |
| 11 | qwen3.5-plusAlibaba | not stated | default | 12.4% | not dated | via Epoch AI |
| 12 | grok-4-20xAI | closed | default | 11.9% | not dated | via Epoch AI |
| 13 | zai-org/GLM-4.7Epoch: zai-org/glm-4.7, Z.ai (Zhipu AI) | open | default | 10.9% | not dated | via Epoch AI |
| 14 | deepseek-ai/DeepSeek-V3.2-ExpEpoch: DeepSeek-V3.2-Exp_thinking, DeepSeek | open | effort thinking | 9.5% | not dated | via Epoch AI |
| 15 | deepseek-ai/DeepSeek-V3.2Epoch: deepseek/deepseek-v3.2, DeepSeek | open | default | 7.4% | not dated | via Epoch AI |
| 16 | mimo-v2-proXiaomi Corp | closed | default | 6.9% | not dated | via Epoch AI |
| 17 | MiniMaxAI/MiniMax-M2.5Epoch: MiniMax-M2.5, MiniMax | open | default | 6.3% | not dated | via Epoch AI |
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