CL-bench Life
Each model’s result as Epoch AI’s Benchmarking Hub publishes it, run by Epoch or collected from the source it cites. Facthory has not run them.
Data: Epoch AI, ‘Capabilities & benchmarking’, epoch.ai/benchmarks, licensed CC BY 4.0. Read on 16 Nov 2026.
| # | Model | Organization | Result | Evidence |
|---|---|---|---|---|
| 1 | GPT-5.5 (high) API access | OpenAI | 22.2% | – |
| 2 | GPT-5.4 (xhigh) API access | OpenAI | 21.7% | – |
| 3 | GPT-5.4 (high) API access | OpenAI | 19.3% | – |
| 4 | GPT-5.1 (high) API access | OpenAI | 17.3% | – |
| 5 | Claude Opus 4.6 (high) API access | Anthropic | 17% | – |
| 6 | Gemini 3.1 Pro Preview API access | Google DeepMind | 16.9% | – |
| 7 | GPT-5.4 (unknown thinking) API access | OpenAI | 13.8% | – |
| 8 | Claude Opus 4.6 (unknown thinking) API access | Anthropic | 13.6% | – |
| 9 | DeepSeek v4 Pro (high) Open weights (unrestricted) | DeepSeek | 13.5% | – |
| 10 | Kimi K2.5 Open weights (unrestricted) | Moonshot | 13.2% | – |
| 11 | Qwen3.5 Plus | Alibaba | 12.4% | – |
| 12 | grok-4-20 API access | xAI | 11.9% | – |
| 13 | GLM-4.7 (Novita) Open weights (unrestricted) | Z.ai (Zhipu AI) | 10.9% | – |
| 14 | DeepSeek-V3.2-Exp_thinking Open weights (unrestricted) | DeepSeek | 9.5% | – |
| 15 | DeepSeek-V3.2 (Thinking; Novita) Open weights (unrestricted) | DeepSeek | 7.4% | – |
| 16 | mimo-v2-pro API access | Xiaomi Corp | 6.9% | – |
| 17 | MiniMax-M2.5 Open weights (unrestricted) | MiniMax | 6.3% | – |
