OpenCompass Reasoning - Abductive - English (CompassBench 2404): leaderboard

Metric: Score (%). Source: rank.opencompass.org.cn. 30 models tracked.

Top models

#ModelScore
1Yi 34B (Chat)86
2GPT-4 Turbo (Preview)84
3OrionStar-Yi-34B-Chat72
4Llama 2 70B Chat70
5Baichuan2-13B-Chat66
6Llama 2 7B Chat54
7Qwen-14B-Chat50
8Llama 2 13B Chat48
9Mistral 7B Instruct (v0.2)46
10Baichuan2-7B-Chat46
11Qwen-72B-Chat42
12Yi 6B (Chat)40
13Qwen-7B Chat40
14WizardLM-13B-V1.232
15WizardLM-70B-V1.028

Interactive version: theaggregate.ai/benchmark?slug=opencompass-reasoning-abductive-english-compassbench-2404 · How It Works · Data refreshed daily, snapshot 2026-09-19.