List your company

Physical Reasoning on PhysicsEval

87.9Pass@1 AccuracyAERO
78.95681.27883.685.922Feb 3, 2026

Evaluation results

MethodLinks
AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R52026.0287.9
AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R42026.0286.9
AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R32026.0286.3
Qwen3-8B-BaseBackbone Model=Qwen3-8B-Base2026.0286.2
AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R22026.0286.1
AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R12026.0285.6
AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R42026.0280.6
AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R32026.0280.3
AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R22026.0280.1
Qwen3-4B-BaseBackbone Model=Qwen3-4B-Base2026.0279.8
AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R12026.0279.5
AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R52026.0279.3