MEGA-Bench Task - Humor Explanation: leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro (Preview 03-25) | 96 |
| 2 | GPT-4o | 86.7 |
| 3 | Gemini 1.5 Flash (002) | 85.3 |
| 4 | GPT-4o Mini | 82 |
| 5 | Gemini 1.5 Pro (002) | 80 |
| 6 | Gemma 3 27B (IT) | 79.3 |
| 7 | Claude 3.5 Sonnet (20241022) | 72.7 |
| 8 | Gemma 3 12B (IT) | 70 |
| 9 | InternVL2.5-78B | 68.7 |
| 10 | Gemma 3 4B (IT) | 66.7 |
| 11 | Claude 3.5 Sonnet (20240620) | 58.7 |
| 12 | InternVL3-8B | 55.3 |
| 13 | InternVL3-78B | 54.7 |
| 14 | Qwen 2 VL 72B | 54.7 |
| 15 | Llama 4 Scout Base | 53.3 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-humor-explanation · How It Works · Data refreshed daily, snapshot 2026-09-05.