List your company

Visual puzzle solving on Jigsaw R1 (test)

61.9Accuracy (2x1)GPT-4.1-mini
44.2248.8153.457.99Aug 7, 2025

Evaluation results

MethodLinks
GPT-4.1-mini2025.0861.954.554.820.347.88
BAGEL2025.0851.2550.0552.059.640.73
Uni-CoT2025.0851.1561.4659.1518.6447.6
Random2025.0850505012.540.63
Qwen2.5-VL-7BParameters=7B2025.0849.448.750.615.841.12
InternVL2.5-2BParameters=2B2025.0844.941.948.69.736.28