ReLE - Overall: leaderboard
ReLE Chinese LLM benchmark (NoneLinear, formerly CLiB), a continuously updated Chinese-language evaluation run through each model's API. Overall score, which ReLE computes as 0.7 x general ability (mean of reasoning and math, language and instruction following, agents and tool use, coding) plus 0.3 x professional ability (mean of education, medicine and mental health, finance, law and civil service). Accuracy 0-100, higher is better.
Metric: Accuracy (%). Source: nonelinear.com. Saturation forecast: Around February 2027. 140 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5.5 | 78.6 | #3 |
| 2 | Gemini 4 Argon | 77.3 | #9 |
| 3 | Qwen 3.7 Max | 76.9 | #85 |
| 4 | Claude Opus 5 | 76.8 | #16 |
| 5 | Kimi K3.1 Pro | 76.7 | #17 |
| 6 | Gemini 3.8 Flash | 76.2 | #30 |
| 7 | Kimi K3 | 75.6 | #35 |
| 8 | Qwen 3.8 Max | 75.4 | #34 |
| 9 | Gemini 3.7 Flash | 75.4 | #24 |
| 10 | GPT-5.5 | 75.3 | #33 |
| 11 | Gemini 3.1 Pro (Preview) | 75.2 | #64 |
| 12 | GLM-5.3 | 75.2 | #44 |
| 13 | DeepSeek V4 Pro | 74.9 | #112 |
| 14 | Hy4 preview | 74.7 | #45 |
| 15 | Claude Opus 4.8 (Thinking) | 74.7 | #39 (Claude Opus 4.8) |
No result here: #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=rele-overall · How It Works · Data refreshed daily, snapshot 2026-11-05.