PuzzleMate - Piece Selection (Aria): leaderboard
A task of PuzzleMate (Dasgupta et al., 2026), an egocentric jigsaw-puzzle benchmark for multimodal assistants built on three 48-piece puzzles: the model picks the piece that fits a missing slot from five candidates marked with numbered boxes, on photos from a low-resolution Project Aria glasses camera. Scored as accuracy in percent; higher is better.
Metric: Accuracy (%; exact answer; task: choosing the piece that fits a missing slot among five candidates marked with numbered boxes; three 48-piece jigsaw puzzles photographed from an egocentric viewpoint, Project Aria glasses camera (low-resolution, noisy egocentric images)). Source: arxiv.org. Saturation forecast: Around August 2028. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 | 50 | #172 |
| 2 | Gemini 2.5 Pro | 41 | #262 |
| 3 | Gemma 3 4B | 34 | #5104 |
| 4 | Gemma 3 12B | 30 | #2013 |
| 5 | Gemma 3 27B | 28 | #1638 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=puzzlemate-piece-selection-aria · How It Works · Data refreshed daily, snapshot 2026-11-12.