Share your thoughts, 1 month free Claude Pro on usSee more

Physics-Scene Visual Reasoning on Physics

54.35Accuracy

S1-VL-32B-RL

Updated 1mo ago

Evaluation Results

Method	Links
S1-VL-32B-RL 2026.04		54.35
GPT-5 2026.04		48.34
Qwen3-VL-235B-A22B-Thinking 2026.04		46.03
Intern-S1 2026.04		44.95
S1-VL-32B-SFT 2026.04		43.8
Qwen3-VL-32B-Thinking 2026.04		41.71
Gemini 2.5 Pro 2026.04		40
Gemini 2.5 Flash 2026.04		37.1
Intern-S1-mini 2026.04		30.53
Thyme-VL 2026.04		12.41