List your company

Physical Reasoning on PHYBench

5.3Pass@1 AccuracyAERO
2.5963.29844.702Feb 3, 2026

Evaluation results

MethodLinks
AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R52026.025.3
AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R42026.025.1
AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R32026.024
AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R52026.023.9
Qwen3-8B-BaseBackbone Model=Qwen3-8B-Base2026.023.8
AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R22026.023.8
AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R42026.023.7
AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R32026.023.4
AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R12026.023.4
Qwen3-4B-BaseBackbone Model=Qwen3-4B-Base2026.022.7
AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R12026.022.7
AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R22026.022.7