OpenCompass Math - Competition (CompassBench 2601): leaderboard
OpenCompass CompassBench 2601 edition (file dated 2026-02-11): competition mathematics sub-task of the math category. 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: Around February 2027. 26 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 (2025-12-11) (High) | 70 | |
| 2 | Qwen 3 Max (2026-01-23) | 70 | #158 |
| 3 | Kimi K2.5 | 69.2 | #152 |
| 4 | Gemini 3 Pro (Preview) | 69 | #71 |
| 5 | Kimi K2 (Thinking) | 66.3 | #252 (Kimi K2) |
| 6 | Claude Opus 4.5 (Thinking) | 65 | #87 (Claude Opus 4.5) |
| 7 | GPT-OSS-120B (High) | 64.5 | #346 (GPT-OSS-120B) |
| 8 | DeepSeek V3.2 | 64.2 | #210 |
| 9 | GLM-4.7 | 64 | #202 |
| 10 | Tencent HY 2.0 Think | 62.9 | #206 |
| 11 | Ring-1T | 61.8 | #326 |
| 12 | Grok 4.1 | 61 | #239 |
| 13 | Claude Sonnet 4.5 (Thinking) | 60.3 | #148 (Claude Sonnet 4.5) |
| 14 | Qwen 3 235B A22B 2507 (Thinking) | 58.7 | #263 |
| 15 | Qwen 3 Next 80B A3B (Thinking) | 58.4 | #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-math-competition-compassbench-2601 · How It Works · Data refreshed daily, snapshot 2026-11-05.