OpenCompass LLM - Code - English (CompassBench 2501): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 40 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O1 Mini (2024-09-12) | 55.1 |
| 2 | DeepSeek R1 | 53.2 |
| 3 | Gemini 2.0 Pro (Preview 02-05) | 52.5 |
| 4 | O1 (2024-12-17) | 51.6 |
| 5 | O3 Mini (2025-01-31) | 51.3 |
| 6 | Qwen 2.5 Max | 50.7 |
| 7 | DeepSeek V3 | 46.1 |
| 8 | GPT-4o (2024-11-20) | 45.2 |
| 9 | Gemini 2.0 Flash (001) | 45.1 |
| 10 | QwQ 32B-Preview | 44.1 |
| 11 | MiniMax-Text-01 | 43.8 |
| 12 | Doubao-1.5-pro-32k-250115 | 42.9 |
| 13 | Claude 3.5 Sonnet (20241022) | 42.3 |
| 14 | Yi Lightning | 41.1 |
| 15 | Mistral Large 2 (Nov) Instruct (2411) | 40.5 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-llm-code-english-compassbench-2501 · How It Works · Data refreshed daily, snapshot 2026-09-19.