AI labsUnited States
LM-Provers models
LM-Provers is an open research team developing language models, datasets, graders, and benchmarks for mathematical theorem proving. Its releases include the QED-Nano proof model and artifacts such as FineProofs, IMOProofBench, ProofBench, and theorem-proof grading benchmarks.
LM-Provers models ranked by capability
Capability is Atlas's single score for each model, fitted to every sourced benchmark result (methodology). Rank is among all 896 ranked models from every lab.
| Model | Released | Capability | Rank | Benchmarks | Weights |
|---|---|---|---|---|---|
| QED-Nano | Feb 12, 2026 | 136.1 | #297 | 14 | Open |