OpenCompass LLM - Language — leaderboard

Metric: Score (%). Source: rank.opencompass.org.cn. 23 models tracked.

Top models

#ModelScore
1GPT-5.4 (High)80.2
2Qwen 3.6 Max Preview79.5
3Kimi K2.675
4Gemini 3.1 Pro (Preview)74.8
5Hy3-preview (High)74.4
6GPT-5.4 Mini (High)73
7GLM-5.171.9
8Qwen 3.5 397B A17B71
9Claude Sonnet 4.6 (High)70.5
10DeepSeek V4 Pro70.1
11Claude Opus 4.7 (High)70
12Qwen 3.6 27B69
13MiniMax-M2.768.1
14DeepSeek V4 Flash66.2
15Gemma 4 31B (IT)64.6

Interactive version: theaggregate.ai/benchmark?slug=opencompass-llm-language · How the rankings work · Data refreshed daily, snapshot 2026-07-22.