Visual Puzzle Solving on VisPuzzle
86AccuracyVisual-OPSD
Evaluation Results
| Method | Links | |
|---|---|---|
| Visual-OPSD3-run average=true, Latency (s)=10.02026.06 | 86 | |
| GPT-53-run average=true2026.06 | 78 | |
| ThinkMorph3-run average=true, Latency (s)=142.82026.06 | 77.5 | |
| Visual-OPSD-Noise3-run average=true, Latency (s)=14.92026.06 | 64.5 | |
| Text-only SFT3-run average=true, Latency (s)=28.52026.06 | 63.5 | |
| Qwen3-VL-32B3-run average=true2026.06 | 50.75 | |
| GPT-4o3-run average=true2026.06 | 43.75 | |
| Qwen3-VL-8B3-run average=true2026.06 | 37 | |
| InternVL3.5-38B3-run average=true2026.06 | 36.5 | |
| BAGEL-7B3-run average=true2026.06 | 35 | |
| InternVL3.5-8B3-run average=true2026.06 | 34.75 | |
| Janus-Pro-7B3-run average=true2026.06 | 33.5 | |
| Chameleon-7B3-run average=true2026.06 | 30.5 |