OpenCompass Code - Execution - English (CompassBench 2507): leaderboard
OpenCompass CompassBench 2507 edition (file dated 2025-08-12): code execution sub-task of the code category, English questions. 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 2032. 28 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Grok 4 | 37.2 | #189 |
| 2 | O4 Mini (2025-04-16) (High) | 31.8 | #183 (O4 Mini (2025-04-16)) |
| 3 | O3 (2025-04-16) (High) | 29.7 | #125 (O3 (2025-04-16)) |
| 4 | Qwen 3 235B A22B 2507 (Thinking) | 29.1 | #263 |
| 5 | DeepSeek V3 (0324) | 29.1 | #349 |
| 6 | GLM-4.5 | 28.4 | #283 |
| 7 | Gemini 2.5 Pro | 27.7 | #159 |
| 8 | DeepSeek R1 0528 | 27.7 | #232 |
| 9 | Hunyuan-T1-20250711 | 27.7 | #277 |
| 10 | GPT-4.1 (2025-04-14) | 27 | #257 |
| 11 | Claude Sonnet 4 (Thinking) | 27 | #207 (Claude Sonnet 4) |
| 12 | Kimi K2 | 26.4 | #252 |
| 13 | Qwen 3 235B A22B 2507 Instruct | 26.4 | #315 |
| 14 | GLM 4.5 Air | 26.4 | #403 |
| 15 | Doubao-Seed-1.6-thinking-250715 | 26.4 | #169 |
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-execution-english-compassbench-2507 · How It Works · Data refreshed daily, snapshot 2026-11-05.