List your company

World Modeling on Plancraft

98.4SmeltNeSyS
28.61646.73364.8582.967Feb 11, 2026

Evaluation results

MethodLinks
NeSySBackbone=Llama3.2-1B, Phase=Phase 2, Training Data=35%2026.0298.495.193.76787.7
GPT-5-miniBackbone=GPT-5-mini2026.0293.891.560.34773.8
NeSySBackbone=Qwen3-4B, Phase=Phase 12026.0293.88392.567.481.6
NeSySBackbone=Qwen3-4B, Phase=Phase 2, Training Data=35%2026.0292.297.196.665.688.4
Llama3.1-8BBackbone=Llama3.1-8B2026.0290.691.269.536.572.6
SFTBackbone=Qwen3-4B, Training Data=100%2026.0287.599.894.87090.1
Neural WMBackbone=Qwen3-4B, Phase=Phase 22026.0287.510093.151.785.1
Symbolic WMBackbone=Qwen3-4B, Phase=Phase 12026.0285.979.382.266.176.9
Symbolic WMBackbone=Llama3.2-1B, Phase=Phase 22026.0281.389.187.962.281.2
Symbolic WMBackbone=Qwen3-4B, Phase=Phase 22026.0281.389.886.845.777
Phi-4-miniBackbone=Phi-4-mini2026.027590.572.43370.8
NeSySBackbone=Llama3.2-1B, Phase=Phase 12026.0271.983.992.569.681
GPT-oss-20BBackbone=GPT-oss-20B2026.0262.581.370.727.463.7
Neural WMBackbone=Qwen3-4B, Phase=Phase 12026.0257.889.369.531.367.9
Qwen3-14BBackbone=Qwen3-14B2026.0256.391.564.435.768.9
Neural WMBackbone=Llama3.2-1B, Phase=Phase 12026.0254.788.169.528.766.4
Qwen3-8BBackbone=Qwen3-8B2026.0248.49068.433.968
Symbolic WMBackbone=Llama3.2-1B, Phase=Phase 12026.0245.393.447.748.769.2
Neural WMBackbone=Llama3.2-1B, Phase=Phase 22026.0234.497.875.34775.4
SFTBackbone=Llama3.2-1B, Training Data=100%2026.0231.398.873.66780.5