Benchmarks
CL Bench Life
Models scored
17
Best score
0.22
Metric
Overall
CL Bench Life is an AI evaluation tracked by Epoch AI, with 17 scored model versions on record. The scores here are reported by third parties rather than produced by Epoch AI running the evaluation itself.
Scored models span September 2025 to April 2026. The best score recorded is 0.22, measured as Overall. Models come from OpenAI, Anthropic, Google DeepMind, DeepSeek, Moonshot, Alibaba, xAI, Z.ai (Zhipu AI) and others.
Full record
- Score metric
- Overall
- Models scored
- 17
- Best score recorded
- 0.22
- Earliest model scored
- Sep 29, 2025
- Most recent model scored
- Apr 24, 2026
- Third-party reported
- Yes
Leaderboard
| 01 | gpt-5.5_high | OpenAI | 0.22 |
| 02 | gpt-5.4-2026-03-05_xhigh | OpenAI | 0.22 |
| 03 | gpt-5.4-2026-03-05_high | OpenAI | 0.19 |
| 04 | gpt-5.1-2025-11-13_high | OpenAI | 0.17 |
| 05 | claude-opus-4-6_high | Anthropic | 0.17 |
| 06 | gemini-3.1-pro-preview | Google DeepMind | 0.17 |
| 07 | gpt-5.4-2026-03-05_unknown | OpenAI | 0.14 |
| 08 | claude-opus-4-6_unknown | Anthropic | 0.14 |
| 09 | deepseek-v4-pro_high | DeepSeek | 0.14 |
| 10 | kimi-k2.5 | Moonshot | 0.13 |
| 11 | qwen3.5-plus | Alibaba | 0.12 |
| 12 | grok-4-20 | xAI | 0.12 |
| 13 | zai-org/glm-4.7 | Z.ai (Zhipu AI) | 0.11 |
| 14 | DeepSeek-V3.2-Exp_thinking | DeepSeek | 0.1 |
| 15 | deepseek/deepseek-v3.2 | DeepSeek | 0.07 |
| 16 | mimo-v2-pro | Xiaomi Corp | 0.07 |
| 17 | MiniMax-M2.5 | MiniMax | 0.06 |
More benchmarks
SourceEpoch AI, 'AI Benchmarking Hub'. Published online at epoch.ai. Retrieved 2026-07-29 from https://epoch.ai/benchmarks. Licensed under CC BY 4.0.
← All benchmarks