List your company

Physical Reasoning on UGPhysics

21.7Pass@1 AccuracyAERO
12.44414.84717.2519.653Feb 3, 2026

Evaluation results

MethodLinks
AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R52026.0221.7
AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R42026.0219.7
AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R52026.0219.4
AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R42026.0218.5
AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R32026.0216.4
AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R32026.0216.2
AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R22026.0215.3
AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R22026.0214.8
AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R12026.0213.2
Qwen3-8B-BaseBackbone Model=Qwen3-8B-Base2026.0213.2
AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R12026.0213.2
Qwen3-4B-BaseBackbone Model=Qwen3-4B-Base2026.0212.8