OpenCompass Code - Competition - Chinese (CompassBench 2507): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 28 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 | 71 |
| 2 | O3 (2025-04-16) (High) | 71 |
| 3 | O4 Mini (2025-04-16) (High) | 63.8 |
| 4 | Hunyuan-T1-20250711 | 59.4 |
| 5 | GLM-4.5 | 58 |
| 6 | DeepSeek R1 0528 | 56.5 |
| 7 | Grok 4 | 55.1 |
| 8 | Qwen 3 235B A22B 2507 (Thinking) | 55.1 |
| 9 | Doubao-Seed-1.6-thinking-250715 | 55.1 |
| 10 | Claude Sonnet 4 (Thinking) | 53.6 |
| 11 | GLM 4.5 Air | 52.2 |
| 12 | Gemini 2.5 Pro | 50.7 |
| 13 | MiniMax M1 80k | 46.4 |
| 14 | Kimi K2 | 39.1 |
| 15 | GPT-4.1 (2025-04-14) | 36.2 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-code-competition-chinese-compassbench-2507 · How It Works · Data refreshed daily, snapshot 2026-09-19.