ESM3 / EvolutionaryScale
Multimodal generative protein language model over sequence, structure and function (ESM3; also ESM C embeddings).
| Category | Protein structure & design |
|---|---|
| GitHub | Biohub/esm |
| First public | 2024-07-02 (source) |
| Paper | https://www.science.org/doi/10.1126/science.ads0018 · Science (2025) |
| License | MIT |
| Pricing | Freemium |
| Tags | #plm#generative |
| Last checked | 2026-10-16 |
Specs
| Task | Generative design, embeddings, structure prediction |
|---|---|
| Modalities | Protein sequence, structure and function tracks |
| MSA required | No (single-sequence generative masked LM) |
| Open weights | ESM3-open: yes (MIT); larger models via API |
| GPU need | CUDA or CPU (esm3-sm-open-v1) |
| GitHub stars | 2,993 |
| Interface | Package |
Specs come from each project's README, paper or official page · ESM3 / EvolutionaryScale README · ESM3 / EvolutionaryScale paper · ESM3 / EvolutionaryScale site · ESM3 / EvolutionaryScale README. “—” means we couldn't verify it.
Compare with…
Pick 1–3 other Protein structure & design tools to compare side by side with ESM3 / EvolutionaryScale.
Related
DeepMind's open-source protein structure prediction model (monomer and multimer).
Structure prediction of proteins, nucleic acids, ligands and their complexes; hosted web server plus open code, weights on request for non-commercial use.
Open biomolecular interaction models (Boltz-1, Boltz-2) for complex structure and binding-affinity prediction.