List your company

Physics Reasoning on Synthetic Symbolic

10.4AccuracyQwen2.5-32B + RL (synthetic)
2.9124.8566.88.744Apr 13, 2026

Evaluation results

MethodLinks
Qwen2.5-32B + RL (synthetic)Model=Qwen2.5-32B, Post-training Stage=RL (synthetic)2026.0410.4
Qwen2.5-14B + RL (synthetic)Model=Qwen2.5-14B, Post-training Stage=RL (synthetic)2026.0410.4
Qwen2.5-7B + RL (synthetic)Model=Qwen2.5-7B, Post-training Stage=RL (synthetic)2026.049.6
Qwen2.5-3B + RL (synthetic)Model=Qwen2.5-3B, Post-training Stage=RL (synthetic)2026.049.4
Qwen3-30BModel=Qwen3-30B, Post-training Stage=Instruct2026.048.8
Qwen3-30B + RL (synthetic)Model=Qwen3-30B, Post-training Stage=RL (synthetic)2026.048
Qwen2.5-32BModel=Qwen2.5-32B, Post-training Stage=Instruct2026.045.6
Qwen2.5-14BModel=Qwen2.5-14B, Post-training Stage=Instruct2026.045.6
Qwen2.5-7BModel=Qwen2.5-7B, Post-training Stage=Instruct2026.045.6
Qwen2.5-3BModel=Qwen2.5-3B, Post-training Stage=Instruct2026.043.2