List your company

Physical Reasoning on Angry Birds

47.06Success RateQwen-7B^P&I GRPO
16.473624.414332.35540.2957Mar 1, 2026

Evaluation results

MethodLinks
Qwen-7B^P&I GRPOPolicy Model=Qwen-7B^P&I GRPO, World Model=Qwen-7B^P&I2026.0347.06
Qwen-7B^P&I GRPOPolicy Model=Qwen-7B^P&I GRPO, World Model=-2026.0345.61
Qwen-7B^P&I SFTPolicy Model=Qwen-7B^P&I SFT, World Model=Qwen-7B^P&I2026.0344.12
Qwen-7B^P GRPOPolicy Model=Qwen-7B^P GRPO, World Model=Qwen-7B^P2026.0344.12
Qwen-7B^P SFTPolicy Model=Qwen-7B^P SFT, World Model=Qwen-7B^P2026.0341.18
Qwen-7B^P&I SFTPolicy Model=Qwen-7B^P&I SFT, World Model=-2026.0341.18
Qwen-7B^P GRPOPolicy Model=Qwen-7B^P GRPO, World Model=-2026.0341.18
Qwen-7B^P SFTPolicy Model=Qwen-7B^P SFT, World Model=-2026.0338.24
Gemini-2.5-ProPolicy Model=Gemini-2.5-Pro, World Model=-2026.0335.29
GPT-o3Policy Model=GPT-o3, World Model=-2026.0335.29
Qwen-3B^P&I GRPOPolicy Model=Qwen-3B^P&I GRPO, World Model=Qwen-3B^P&I2026.0335.29
Claude 4.0 OpusPolicy Model=Claude 4.0 Opus, World Model=-2026.0332.35
Qwen-3B^P&I SFTPolicy Model=Qwen-3B^P&I SFT, World Model=Qwen-3B^P&I2026.0332.35
Qwen-72BPolicy Model=Qwen-72B, World Model=-2026.0329.41
Claude 4.0 OpusPolicy Model=Claude 4.0 Opus, World Model=Claude 4.0 Opus2026.0329.41
Gemini-2.5-ProPolicy Model=Gemini-2.5-Pro, World Model=Gemini-2.5-Pro2026.0329.41
GPT-o3Policy Model=GPT-o3, World Model=GPT-o32026.0329.41
Qwen-3B^P GRPOPolicy Model=Qwen-3B^P GRPO, World Model=Qwen-3B^P2026.0329.41
Qwen-3B^P&I GRPOPolicy Model=Qwen-3B^P&I GRPO, World Model=-2026.0329.41
Qwen-32BPolicy Model=Qwen-32B, World Model=-2026.0326.47
Qwen-72BPolicy Model=Qwen-72B, World Model=Qwen-72B2026.0326.47
Qwen-3B^P SFTPolicy Model=Qwen-3B^P SFT, World Model=Qwen-3B^P2026.0326.47
Qwen-3B^P&I SFTPolicy Model=Qwen-3B^P&I SFT, World Model=-2026.0326.47
Qwen-3B^P GRPOPolicy Model=Qwen-3B^P GRPO, World Model=-2026.0326.47
Qwen-7B^I SFTPolicy Model=Qwen-7B^I SFT, World Model=Qwen-7B^I2026.0326.47
Qwen-7B^I GRPOPolicy Model=Qwen-7B^I GRPO, World Model=Qwen-7B^I2026.0326.47
Qwen-3B^P SFTPolicy Model=Qwen-3B^P SFT, World Model=-2026.0323.53
Qwen-3B^I SFTPolicy Model=Qwen-3B^I SFT, World Model=Qwen-3B^I2026.0323.53
Qwen-3B^I GRPOPolicy Model=Qwen-3B^I GRPO, World Model=Qwen-3B^I2026.0323.53
Qwen-7B^I SFTPolicy Model=Qwen-7B^I SFT, World Model=-2026.0323.53
Qwen-7B^I GRPOPolicy Model=Qwen-7B^I GRPO, World Model=-2026.0323.53
Qwen-7BPolicy Model=Qwen-7B, World Model=-2026.0320.59
Qwen-3B^I SFTPolicy Model=Qwen-3B^I SFT, World Model=-2026.0320.59
Qwen-3B^I GRPOPolicy Model=Qwen-3B^I GRPO, World Model=-2026.0320.59
MOCKPolicy Model=MOCK, World Model=-2026.0317.65
Qwen-3BPolicy Model=Qwen-3B, World Model=-2026.0317.65