MEGA-Bench Task - Number Comparison — leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 100 |
| 2 | GPT-4o | 100 |
| 3 | Gemma 3 27B (IT) | 100 |
| 4 | InternVL3-8B | 100 |
| 5 | Qwen 2 VL 72B | 100 |
| 6 | Claude 3.5 Sonnet (20241022) | 100 |
| 7 | Ovis2-8B | 100 |
| 8 | Pixtral-12B | 100 |
| 9 | InternVL2.5-78B | 100 |
| 10 | Claude 3.5 Sonnet (20240620) | 100 |
| 11 | Llama 4 Scout Base | 100 |
| 12 | Gemini 2.0 Flash (Preview) | 100 |
| 13 | Gemma 3 12B (IT) | 92.9 |
| 14 | Qwen 2 VL 2B | 92.9 |
| 15 | InternVL3-14B | 92.9 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-number-comparison · How the rankings work · Data refreshed daily, snapshot 2026-07-22.