OpenCompass Code - Competition — leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 26 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 73.2 |
| 2 | Qwen 3 Max (2026-01-23) | 66.1 |
| 3 | GPT-5.2 (High) | 62.9 |
| 4 | GPT-OSS-120B (High) | 61.9 |
| 5 | DeepSeek V3.2 | 59.8 |
| 6 | GLM-4.7 | 59 |
| 7 | Claude Opus 4.5 (Thinking) | 58.7 |
| 8 | Kimi K2.5 | 58.2 |
| 9 | Kimi K2 (Thinking) | 56.6 |
| 10 | MiniMax-M2.1 | 53.5 |
| 11 | O4 Mini (2025-04-16) (High) | 48.3 |
| 12 | Qwen 3 235B A22B (Thinking) | 46.4 |
| 13 | Ring-1T | 42.9 |
| 14 | Grok 4.1 | 42.7 |
| 15 | Qwen 3 Next 80B A3B (Thinking) | 42 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-code-competition · How the rankings work · Data refreshed daily, snapshot 2026-07-22.