Capability benchmark · Chemistry · Materials science
OMol25 scales open molecular DFT data for chemistry models
OMol25 scales open molecular DFT data for chemistry models: capability signal for AI systems on research-adjacent tasks.
Summary
The arXiv paper introduces Open Molecules 2025, a Meta FAIR dataset with more than 100 million DFT calculations and about 83 million unique molecular systems spanning small molecules, biomolecules, metal complexes, electrolytes, charges, spins and reactive structures.
AI role
AI systems are tested on research-adjacent capabilities relevant to chemistry, materials science.
Narrative role
This is supporting evidence for whether AI systems can perform research-adjacent tasks needed before stronger discovery or acceleration claims.
Caveat
Benchmark, model, or tool performance is an upstream capability indicator, not proof of new scientific discovery.