Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Interphyre

Benchmarks

Task NameDataset NameSOTA ResultTrend
Physics-based puzzle solvingINTERPHYRE 1.0 (test)
Cliffhanger Score100
12
Physical ReasoningINTERPHYRE (test)
Cliffhanger0
5
Physical ReasoningInterPhyre Two Body Problem (evaluation)
Accuracy34
4
Physical ReasoningInterPhyre Down to Earth (evaluation)
Accuracy72
4
Physics Problem SolvingInterphyre Two Body Problem level 1.0 (test)
Accuracy34
4
Physics Problem SolvingInterphyre Down to Earth level 1.0 (test)
Accuracy72
4
Physical ReasoningInterPhyre Catapult (evaluation)
Accuracy54
2
Physics Problem SolvingInterphyre Catapult level 1.0 (test)
Accuracy54
2
Showing 8 of 8 rows