OpenCompass Language - Dialogue (CompassBench 2601): leaderboard
OpenCompass CompassBench 2601 edition (file dated 2026-02-11): dialogue sub-task of the language category. 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. 26 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 Max (2026-01-23) | 98.9 | #158 |
| 2 | Claude Sonnet 4.5 (Thinking) | 98.8 | #148 (Claude Sonnet 4.5) |
| 3 | Kimi K2.5 | 98.6 | #152 |
| 4 | Kimi K2 (Thinking) | 98.5 | #252 (Kimi K2) |
| 5 | Qwen 3 235B A22B 2507 (Thinking) | 98.1 | #263 |
| 6 | Claude Opus 4.5 (Thinking) | 98.1 | #87 (Claude Opus 4.5) |
| 7 | GPT-OSS-120B (High) | 98 | #346 (GPT-OSS-120B) |
| 8 | Ring-1T | 98 | #326 |
| 9 | MiniMax-M2.1 | 97 | #299 |
| 10 | Doubao-Seed-1-8-251228 | 96.4 | #153 |
| 11 | Ling-1T | 96.3 | #354 |
| 12 | Grok 4.1 | 96.2 | #239 |
| 13 | O3 (2025-04-16) (High) | 96.1 | #125 (O3 (2025-04-16)) |
| 14 | Qwen 3 Next 80B A3B (Thinking) | 95.5 | #312 |
| 15 | Seed-OSS-36B-Instruct | 94.4 | #384 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=opencompass-language-dialogue-compassbench-2601 · How It Works · Data refreshed daily, snapshot 2026-11-05.