ReLE - Reasoning and Mathematics: leaderboard
ReLE Chinese LLM benchmark (NoneLinear, formerly CLiB), a continuously updated Chinese-language evaluation run through each model's API. Reasoning and mathematical calculation domain: deductive and commonsense reasoning, BBH, arithmetic, table QA and summarization, olympiad math and sudoku items. Accuracy 0-100, higher is better.
Metric: Accuracy (%). Source: nonelinear.com. Saturation forecast: Around February 2027. 145 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5 | 92 | #16 |
| 2 | Claude Opus 5.5 | 90.6 | #3 |
| 3 | Claude Opus 4.8 (Thinking) | 89.9 | #39 (Claude Opus 4.8) |
| 4 | Kimi K3.1 Pro | 87.9 | #17 |
| 5 | Qwen 3.8 Flash | 87.7 | #62 |
| 6 | Kimi K3 | 86.3 | #35 |
| 7 | Claude Sonnet 5.5 | 85.7 | #14 |
| 8 | Qwen 3.5 122B A10B | 85.5 | #184 |
| 9 | Gemini 4 Argon | 85.4 | #9 |
| 10 | Gemini 3.1 Pro (Preview) | 85.1 | #64 |
| 11 | GLM-5.3 Flash | 84.8 | #77 |
| 12 | GPT-5.2 (High) | 84.8 | #115 (GPT-5.2) |
| 13 | Qwen 3.7 Max | 84.7 | #85 |
| 14 | GPT-5.1 (High) | 84.7 | #145 (GPT-5.1) |
| 15 | Gemini 3.5 Flash | 84.5 | #66 |
No result here: #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=rele-reasoning-and-mathematics · How It Works · Data refreshed daily, snapshot 2026-11-05.