OpenCompass LLM - Reasoning - English (CompassBench 2405): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 39 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Yi Large | 45.9 |
| 2 | GPT-4 Turbo | 45.5 |
| 3 | GPT-4o (2024-05-13) | 43.2 |
| 4 | Claude 3 Opus | 42.2 |
| 5 | Yi-1.5-9B Chat | 41.3 |
| 6 | Llama 3 70B Instruct | 39.7 |
| 7 | Yi 1.5 34B Chat | 39 |
| 8 | internlm2-chat-7B | 35 |
| 9 | GLM-4 | 34.3 |
| 10 | DeepSeek V2 Chat | 34 |
| 11 | Qwen 1.5 32B Chat | 33.6 |
| 12 | Moonshot v1 8K | 32.9 |
| 13 | Qwen 1.5 72B Chat | 32.7 |
| 14 | Mixtral 8x22B Instruct (v0.1) | 30.9 |
| 15 | Command-R+ | 30.8 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-llm-reasoning-english-compassbench-2405 · How It Works · Data refreshed daily, snapshot 2026-09-19.