OpenCompass Code - MBPP - English (CompassBench 2405): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 38 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o (2024-05-13) | 87.2 |
| 2 | Yi Large | 79 |
| 3 | GPT-4 Turbo | 77.4 |
| 4 | GPT-3.5 Turbo | 77 |
| 5 | Llama 3 70B Instruct | 75.9 |
| 6 | Claude 3 Opus | 75.5 |
| 7 | Mixtral 8x22B Instruct (v0.1) | 75.5 |
| 8 | GLM-4 | 74.3 |
| 9 | Yi-1.5-9B Chat | 73.5 |
| 10 | DeepSeek V2 Chat | 73.2 |
| 11 | Yi 1.5 34B Chat | 72.8 |
| 12 | Yi-1.5-6B-Chat | 69.3 |
| 13 | Moonshot v1 8K | 67.3 |
| 14 | Command-R+ | 66.2 |
| 15 | internlm2-chat-20B | 65.8 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-code-mbpp-english-compassbench-2405 · How It Works · Data refreshed daily, snapshot 2026-09-19.