OpenCompass Reasoning - Abductive - English (CompassBench 2405): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 39 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3 Opus | 92 |
| 2 | Yi-1.5-9B Chat | 88 |
| 3 | GPT-4 Turbo | 86 |
| 4 | Yi Large | 86 |
| 5 | Llama 3 70B Instruct | 82 |
| 6 | GPT-4o (2024-05-13) | 80 |
| 7 | Moonshot v1 8K | 80 |
| 8 | Yi 1.5 34B Chat | 76 |
| 9 | GLM-4 | 74 |
| 10 | Mixtral 8x22B Instruct (v0.1) | 72 |
| 11 | DeepSeek V2 Chat | 72 |
| 12 | Qwen 1.5 32B Chat | 66 |
| 13 | Command-R+ | 64 |
| 14 | internlm2-chat-20B | 62 |
| 15 | internlm2-chat-7B | 62 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-reasoning-abductive-english-compassbench-2405 · How It Works · Data refreshed daily, snapshot 2026-09-19.