SuperCLUE General (October 2024) - Hard: leaderboard
Metric: Score. Source: www.superclueai.com. 42 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O1 Preview | 64.89 |
| 2 | Claude 3.5 Sonnet (20241022) | 53.6 |
| 3 | GLM-4 Plus | 51.09 |
| 4 | Gemini 1.5 Pro (002) | 47.82 |
| 5 | GPT-4 Turbo | 46.82 |
| 6 | Qwen 2.5 72B Instruct | 46.64 |
| 7 | Claude 3.5 Sonnet (20240620) | 46.49 |
| 8 | Llama 3.1 405B Instruct | 45.46 |
| 9 | Step 2 16K | 43.91 |
| 10 | 360gpt2-pro | 43.68 |
| 11 | DeepSeek V2.5 | 43.5 |
| 12 | GPT-4o Mini | 41.46 |
| 13 | Llama 3.1 70B Instruct | 37.71 |
| 14 | Llama 3.2 90B Instruct | 36.84 |
| 15 | Qwen 2.5 7B Instruct | 33.92 |
Interactive version: theaggregate.ai/benchmark?slug=superclue-general-october-2024-hard · How It Works · Data refreshed daily, snapshot 2026-09-19.