Benchmark Radar™
Star —

PhysicsBench

Explore score history ↗

PhysicsBench is a unified benchmark and leaderboard for generative and predictive AI models in engineering design and simulation. It spans seven tasks across 1D, 2D, and 3D domains, ranks 66 models on nine datasets, and uses a common metric suite including BenchRank for debiased ranking. Evaluation covers data scales from S to XL.

Released
2026-08-25
Institutions
Korea Advanced Institute of Science and Technology
Evaluates
General AI, Language & Knowledge, cs.LG
Openness
unknown
Importer
Claire Radar
Review status
ai-reviewed

PhysicsBench paper and code

No reported scores are on record for this benchmark yet.

Which sources cite PhysicsBench?

Related General AI, Language & Knowledge, cs.LG benchmarks