OpenCompass LLM - Code (CompassBench 2507): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 28 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O3 (2025-04-16) (High) | 50.6 |
| 2 | O4 Mini (2025-04-16) (High) | 48.1 |
| 3 | GPT-5 | 47.9 |
| 4 | Grok 4 | 45.7 |
| 5 | Hunyuan-T1-20250711 | 43.8 |
| 6 | DeepSeek R1 0528 | 42.9 |
| 7 | Qwen 3 235B A22B 2507 (Thinking) | 41.6 |
| 8 | GLM 4.5 Air | 41.1 |
| 9 | Doubao-Seed-1.6-thinking-250715 | 40.4 |
| 10 | Gemini 2.5 Pro | 39.9 |
| 11 | GLM-4.5 | 39.6 |
| 12 | Claude Sonnet 4 (Thinking) | 38.2 |
| 13 | MiniMax M1 80k | 37.6 |
| 14 | DeepSeek V3 (0324) | 33.1 |
| 15 | Kimi K2 | 32.6 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-llm-code-compassbench-2507 · How It Works · Data refreshed daily, snapshot 2026-09-19.