OpenCompass LLM - Instruction Following - English (CompassBench 2501): leaderboard

Metric: Score (%). Source: rank.opencompass.org.cn. 40 models tracked.

Top models

#ModelScore
1GPT-4o (2024-11-20)68.8
2MiniMax-Text-0167.3
3DeepSeek R162.7
4Gemini 2.0 Pro (Preview 02-05)57.3
5Yi Lightning57.3
6GLM-4 Plus57.1
7Mistral Large 2 (Nov) Instruct (2411)56
8DeepSeek V355
9GPT-4o Mini (2024-07-18)53.3
10Llama 3.1 70B Instruct51.7
11Gemini 2.0 Flash (001)50.4
12Qwen 2.5 Max48.3
13QwQ 32B-Preview46.7
14Llama 3.1 405B Instruct FP846
15Yi 1.5 34B Chat45.8

Interactive version: theaggregate.ai/benchmark?slug=opencompass-llm-instruction-following-english-compassbench-2501 · How It Works · Data refreshed daily, snapshot 2026-09-19.