OpenCompass LLM - Language (CompassBench 2507): leaderboard
OpenCompass CompassBench 2507 edition (file dated 2025-08-12): language category score. CompassBench is OpenCompass's periodic evaluation of chat models on a private question set; each edition is a separate board. Score 0-100, higher is better.
Metric: Score (%). Source: rank.opencompass.org.cn. Saturation forecast: Estimated already saturated. 28 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Doubao-Seed-1.6-thinking-250715 | 86.8 | #169 |
| 2 | Qwen 3 235B A22B 2507 (Thinking) | 84.9 | #263 |
| 3 | Qwen 3 235B A22B 2507 Instruct | 84.3 | #315 |
| 4 | DeepSeek R1 0528 | 84.3 | #232 |
| 5 | GPT-5 | 83.1 | #105 |
| 6 | Grok 4 | 81.3 | #189 |
| 7 | Gemini 2.5 Pro | 79.9 | #159 |
| 8 | ERNIE-X1-Turbo-32K | 79.5 | #288 |
| 9 | Hunyuan-T1-20250711 | 79.2 | #277 |
| 10 | Kimi K2 | 78.4 | #252 |
| 11 | O3 (2025-04-16) (High) | 78.4 | #125 (O3 (2025-04-16)) |
| 12 | GLM-4.5 | 78.2 | #283 |
| 13 | Gemma 3 27B (IT) | 77.1 | #528 |
| 14 | O4 Mini (2025-04-16) (High) | 75.8 | #183 (O4 Mini (2025-04-16)) |
| 15 | GLM 4.5 Air | 75.4 | #403 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=opencompass-llm-language-compassbench-2507 · How It Works · Data refreshed daily, snapshot 2026-11-05.