OpenCompass Code - Competition (CompassBench 2601): leaderboard
OpenCompass CompassBench 2601 edition (a fresh private question set per edition), competition coding subset of the code category. Score 0-100; higher is better. Pinned to the 26-01 edition because the newest CompassBench edition no longer publishes this table.
Metric: Score (%). Source: rank.opencompass.org.cn. Saturation forecast: Not forecast. 26 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 73.2 | #71 |
| 2 | Qwen 3 Max (2026-01-23) | 66.1 | #158 |
| 3 | GPT-5.2 (2025-12-11) (High) | 62.9 | |
| 4 | GPT-OSS-120B (High) | 61.9 | #346 (GPT-OSS-120B) |
| 5 | DeepSeek V3.2 | 59.8 | #210 |
| 6 | GLM-4.7 | 59 | #202 |
| 7 | Claude Opus 4.5 (Thinking) | 58.7 | #87 (Claude Opus 4.5) |
| 8 | Kimi K2.5 | 58.2 | #152 |
| 9 | Kimi K2 (Thinking) | 56.6 | #252 (Kimi K2) |
| 10 | MiniMax-M2.1 | 53.5 | #299 |
| 11 | O4 Mini (2025-04-16) (High) | 48.3 | #183 (O4 Mini (2025-04-16)) |
| 12 | Qwen 3 235B A22B 2507 (Thinking) | 46.4 | #263 |
| 13 | Ring-1T | 42.9 | #326 |
| 14 | Grok 4.1 | 42.7 | #239 |
| 15 | Qwen 3 Next 80B A3B (Thinking) | 42 | #312 |
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-competition-compassbench-2601 · How It Works · Data refreshed daily, snapshot 2026-11-05.