SuperCLUE General (December 2024) - Hard: leaderboard
SuperCLUE general leaderboard, December 2024 edition (2024年12月): Hard group score covering deep reasoning, instruction following and agent tasks (总排行榜 sheet). SuperCLUE's periodic Chinese-language general evaluation uses a new question set each edition, so editions are separate boards. Score 0-100, higher is better.
Metric: Score. Source: www.superclueai.com. Saturation forecast: Estimated already saturated. 38 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O1 | 76.7 | #190 |
| 2 | O1 Preview | 63.6 | #205 |
| 3 | Gemini 2.0 Flash (Preview) | 55.5 | #376 |
| 4 | DeepSeek V3 | 54.8 | #329 |
| 5 | Claude 3.5 Sonnet (20241022) | 54.6 | #302 |
| 6 | Step 2 16K | 50 | #442 |
| 7 | Qwen 2.5 72B Instruct | 49.7 | #469 |
| 8 | Grok 2 (1212) | 49.2 | #502 |
| 9 | GLM-4 Plus | 48.5 | #398 |
| 10 | DeepSeek V2.5 | 45.3 | #494 |
| 11 | Qwen 2.5 32B Instruct | 44.9 | #509 |
| 12 | GPT-4o Mini | 42.8 | #598 |
| 13 | Llama 3.3 70B Instruct | 38.8 | #541 |
| 14 | Qwen 2.5 7B Instruct | 35.7 | #863 |
| 15 | GLM-4 9B Chat | 31.6 | #911 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=superclue-general-december-2024-hard · How It Works · Data refreshed daily, snapshot 2026-11-05.