SuiChat-CN: leaderboard
Metric: Standard F1 (self-reported). Source: benchmarklist.com. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Kimi K2 | 76.3 |
| 2 | Gemini 2.5 Pro | 75.9 |
| 3 | Seed-OSS-36B-Instruct | 74.5 |
| 4 | Qwen 3.5 Plus | 74.2 |
| 5 | Llama 3.3 70B | 71.6 |
| 6 | DeepSeek R1 | 71.5 |
| 7 | Kimi K2.5 | 71.1 |
| 8 | GPT-5 | 70.7 |
| 9 | DeepSeek V3 | 70.6 |
| 10 | Qwen 3 235B A22B | 70.5 |
| 11 | Qwen 3 32B | 70.2 |
| 12 | GLM-5.1 | 69.5 |
| 13 | Qwen 3.5 397B A17B | 69.5 |
| 14 | Kimi K2.6 | 69.4 |
| 15 | GLM-4.6 | 69.1 |
Interactive version: theaggregate.ai/benchmark?slug=suichat-cn · How It Works · Data refreshed daily, snapshot 2026-09-05.