SuperCLUE General (September 2025) - Science Reasoning: leaderboard
Metric: Score. Source: www.superclueai.com. 37 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Kimi K2 (Thinking) | 54.31 |
| 2 | O4 Mini (High) | 54.31 |
| 3 | Gemini 3 Pro (Preview) | 53.45 |
| 4 | GPT-5.1 (High) | 53.45 |
| 5 | Doubao-Seed-1.6-thinking-250715 | 52.59 |
| 6 | GPT-OSS-120B | 50.86 |
| 7 | Gemini 2.5 Pro | 50 |
| 8 | Qwen 3 Max | 48.28 |
| 9 | DeepSeek V3.1 Terminus (Thinking) | 47.41 |
| 10 | Qwen 3 235B A22B 2507 (Thinking) | 46.55 |
| 11 | DeepSeek V3.2 Exp (Thinking) | 45.69 |
| 12 | Claude Opus 4.1 (Thinking) | 45.69 |
| 13 | GPT-5 (High) | 44.83 |
| 14 | GLM-4.6 | 43.97 |
| 15 | GLM-4.5 | 43.1 |
Interactive version: theaggregate.ai/benchmark?slug=superclue-general-september-2025-science-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-19.