MEGA-Bench Task - 2D Image Jigsaw Puzzle Easy: leaderboard
The "2D Image Jigsaw Puzzle Easy" subtask of MEGA-Bench Task. A single task from MEGA-Bench, a large-scale multimodal evaluation suite that scores vision-language models across 500+ real-world tasks spanning perception, reasoning, OCR, charts, documents, and multi-image understanding.
Metric: Task Score (%). Source: huggingface.co. Saturation forecast: Around 2028. 44 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o Mini | 38.1 | #599 |
| 2 | InternVL3-14B | 37.7 | #505 |
| 3 | Claude 3.5 Sonnet (20240620) | 34.3 | #299 |
| 4 | Gemini 2.5 Pro (Preview 03-25) | 33.8 | #123 |
| 5 | GPT-4o | 31.1 | #343 |
| 6 | Gemma 3 27B (IT) | 30.6 | #523 |
| 7 | Claude 3.5 Sonnet (20241022) | 29.3 | #291 |
| 8 | InternVL3-38B | 27.4 | #405 |
| 9 | Llama 4 Scout Base | 26.9 | #707 |
| 10 | Gemma 3 12B (IT) | 25.7 | #664 |
| 11 | Qwen 2 VL 72B | 25.5 | #460 |
| 12 | Gemini 1.5 Pro (002) | 21.3 | #349 |
| 13 | InternVL2.5-78B | 21.2 | #431 |
| 14 | InternVL3-8B | 21 | #620 |
| 15 | InternVL2-8B | 20.6 | #817 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-2d-image-jigsaw-puzzle-easy · How It Works · Data refreshed daily, snapshot 2026-10-20.