OpenCompass LLM - Code (CompassBench 2405): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 39 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o (2024-05-13) | 65.3 |
| 2 | GPT-4 Turbo | 61.5 |
| 3 | DeepSeek V2 Chat | 57.1 |
| 4 | Claude 3 Opus | 55.2 |
| 5 | Yi Large | 50.7 |
| 6 | Mixtral 8x22B Instruct (v0.1) | 50 |
| 7 | Moonshot v1 8K | 48 |
| 8 | GPT-3.5 Turbo | 47.8 |
| 9 | Yi-1.5-9B Chat | 47.6 |
| 10 | Llama 3 70B Instruct | 46.5 |
| 11 | Yi 1.5 34B Chat | 46 |
| 12 | Qwen 1.5 110B Chat | 45.4 |
| 13 | GLM-4 | 44.5 |
| 14 | Yi-1.5-6B-Chat | 41.4 |
| 15 | internlm2-chat-20B | 39.5 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-llm-code-compassbench-2405 · How It Works · Data refreshed daily, snapshot 2026-09-19.