List your company

Environment Generation on LogicEnvEval

100Physics Pass Rate (Floor Plan)Holodeck
59.398469.939280.4891.0208Jan 20, 2026

Evaluation results

MethodLinks
HolodeckModel=DeepSeek-v32026.0110010010010037.057624243226.6715.67
LogicEnvGenModel=DeepSeek-v32026.0110010010010099.0693.7596929694.6749.88
HolodeckModel=Gemini-2.5-Flash2026.0110010010010037.056832242025.3317.88
LogicEnvGenModel=Gemini-2.5-Flash2026.0110010010010096.4792.7896929293.3350.96
HolodeckModel=Qwen2.5-72B2026.0110010010010037.056416121614.6713.96
LogicEnvGenModel=Qwen2.5-72B2026.0110010010010094.7980.1884849286.6753.36
CoTModel=DeepSeek-v32026.019640205263.3885.7164565658.6729.16
IFGModel=DeepSeek-v32026.0190.841113.0438.2991.0890.7292928890.6787.96
CoTModel=Gemini-2.5-Flash2026.0186.4604.1730.2186.179.788488688055.2
IFGModel=Qwen2.5-72B2026.0185.1618.5225.4843.0592.6475.28476928491.64
CoTModel=Qwen2.5-72B2026.018416264265.4773.776056525637.6
IFGModel=Gemini-2.5-Flash2026.0160.9614.1617.4430.8596.1189.09100928893.3389.52