OpenCompass LLM - Instruction Following (CompassBench 2504): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 28 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | ERNIE-X1-Turbo-32K | 71.5 |
| 2 | DeepSeek R1 | 70.1 |
| 3 | Doubao-1.5-thinking-pro-250415 | 69.4 |
| 4 | Claude 3.7 Sonnet (Thinking) | 66.2 |
| 5 | Qwen 3 235B A22B (Thinking) | 65.4 |
| 6 | Gemini 2.5 Pro (Preview 05-06) | 63.8 |
| 7 | Gemma 3 27B (IT) | 62.6 |
| 8 | Qwen 3 32B (Thinking) | 60 |
| 9 | DeepSeek V3 (0324) | 58.5 |
| 10 | Qwen 3 235B A22B | 55.8 |
| 11 | Grok 3 | 54.8 |
| 12 | O4 Mini (High) | 54.1 |
| 13 | GPT-4.1 Mini | 54 |
| 14 | MiniMax-Text-01 | 52.6 |
| 15 | GPT-4.1 (2025-04-14) | 51.2 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-llm-instruction-following-compassbench-2504 · How It Works · Data refreshed daily, snapshot 2026-09-19.