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

Multiple-Choice Question Answering on PhyX 3k (test)

84.2Exact Match Accuracy

Qwen3-VL-32B-Thinking

Updated 19d ago

Evaluation Results

Method	Links
Qwen3-VL-32B-Thinking 2026.05		84.2
Claude Sonnet 4.5 2026.05		80.6
Physics-R1 (dense) 2026.05		77.5
Physics-R1 (binary, seed 42) 2026.05		76.9
Physics-R1 (binary, 3-seed mean ±σ) 2026.05		76.9
Qwen3-VL-8B-Thinking (base) 2026.05		74.4
GPT-4o 2026.05		53.6
Gemini 2.5 Pro 2026.05		49.8
InternVL3-8B 2026.05		43.1