Cl Bench benchmark scores from Epoch AI's benchmark data collection.
Sourced from Epoch AI's systematic benchmark data collection covering frontier models.
Auto-generated definition — editorial description not yet written.
Higher is better.
Auto-generated definition — editorial description not yet written.
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 (xhigh) | 0.3 |
| 2 | GPT-5.1 (high) | 0.2 |
| 3 | Grok 4.20 0309 v2 (Reasoning) | 0.2 |
| 4 | Claude Opus 4.5 (Non-reasoning) | 0.2 |
| 5 | Gemini 3.1 Pro Preview | 0.2 |
| 6 | Qwen3.6 Plus | 0.2 |
| 7 | Qwen: Qwen3.5 Plus 2026-02-15 | 0.2 |
| 8 | Kimi K2.5 (Reasoning) | 0.2 |
| 9 | GLM-5 (Reasoning) | 0.2 |
| 10 | o3 | 0.2 |
| 11 | GLM-4.7 (Reasoning) | 0.2 |
| 12 | Gemini 3 Pro Preview (low) | 0.2 |
| 13 | MiMo-V2-Pro | 0.2 |
| 14 | Qwen3 Max | 0.1 |
| 15 | DeepSeek V3.2 (Non-reasoning) | 0.1 |
| 16 | Kimi K2 Thinking | 0.1 |
| 17 | MiniMax-M2.5 | 0.1 |