OpenCompass Code - Code QA - Chinese (CompassBench 2501): leaderboard
OpenCompass CompassBench 2501 edition (file dated 2025-02-20): code question answering sub-task of the code category, Chinese questions. CompassBench is OpenCompass's periodic evaluation of chat models on a private question set; each edition is a separate board. Score 0-100, higher is better.
Metric: Score (%). Source: rank.opencompass.org.cn. Saturation forecast: Estimated already saturated. 40 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O1 Mini (2024-09-12) | 81.9 | #412 |
| 2 | MiniMax-Text-01 | 80.6 | #519 |
| 3 | DeepSeek R1 | 79.3 | #262 |
| 4 | Gemini 2.0 Pro (Preview 02-05) | 77.2 | #318 |
| 5 | Claude 3.5 Sonnet (20241022) | 75.4 | #302 |
| 6 | Qwen 2.5 Max | 74.6 | #427 |
| 7 | Gemini 2.0 Flash (001) | 74.4 | #365 |
| 8 | O3 Mini (2025-01-31) | 73 | #249 |
| 9 | O1 (2024-12-17) | 72 | #157 |
| 10 | GPT-4o (2024-11-20) | 69.8 | #392 |
| 11 | Doubao-1.5-pro-32k-250115 | 68.3 | #320 |
| 12 | Qwen 2.5 72B Instruct | 63.4 | #469 |
| 13 | Yi Lightning | 62.1 | #429 |
| 14 | DeepSeek V3 | 61.2 | #329 |
| 15 | GLM-Zero-Preview | 60.3 | #682 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=opencompass-code-code-qa-chinese-compassbench-2501 · How It Works · Data refreshed daily, snapshot 2026-11-05.