SuperCLUE General (March 2025) - Text Understanding and Creation: leaderboard
Metric: Score. Source: www.superclueai.com. 36 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | QwQ-32B | 82.27 |
| 2 | Gemini 2.0 Flash (01-21) (Thinking) | 81.16 |
| 3 | DeepSeek R1 | 80.41 |
| 4 | GPT-4.5 (Preview) | 80.4 |
| 5 | Step 2 16K | 80.17 |
| 6 | Gemini 2.0 Pro (Preview 02-05) | 79.34 |
| 7 | GLM-4 Plus | 79.17 |
| 8 | O3 Mini (High) | 79.01 |
| 9 | DeepSeek V3 | 78.99 |
| 10 | hunyuan-turbos-20250226 | 78.88 |
| 11 | Claude 3.7 Sonnet | 78.77 |
| 12 | Qwen 2.5 72B Instruct | 78.52 |
| 13 | Doubao-1.5-pro-32k-250115 | 77.66 |
| 14 | DeepSeek R1 Distill Qwen 32B | 77.53 |
| 15 | Qwen 2.5 14B Instruct | 76.87 |
Interactive version: theaggregate.ai/benchmark?slug=superclue-general-march-2025-text-understanding-and-creation · How It Works · Data refreshed daily, snapshot 2026-09-19.