{
 "schema_version": 1,
 "generated": "2026-10-31",
 "repo": "Biohub/esm",
 "name": "esm",
 "owner": "Biohub",
 "slug": "biohub-esm",
 "github": "https://github.com/Biohub/esm",
 "what": "The esm package from Biohub (EvolutionaryScale) ships ESMC, a protein language model trained on billions of protein sequences, ESMFold2, a structure prediction model built on ESMC 6B that predicts all atom 3D structures from sequence, and sparse autoencoders that decompose ESMC representations into interpretable features. Models run locally from Hugging Face weights or through the Biohub Platform with an API token.",
 "why": "3k stars, picked up by one AI source we track",
 "language": "",
 "requirements": [
  "Python with pip; `pip install esm` is about 5.6 GB, of which about 4.6 GB is the CUDA build of torch",
  "A CUDA GPU for the local examples, which load models with device set to cuda",
  "A Biohub Platform API token if you use the hosted client instead of local weights"
 ],
 "audience_note": "Computational biologists and ML engineers who want protein embeddings, structure predictions or interpretable features without building the stack themselves.",
 "risk": "none",
 "section": "earlier",
 "weekly_rank": null,
 "category": "Security and science",
 "url": "https://dedupedai.com/use-cases/biohub-esm/",
 "use_case_count": 3,
 "use_cases": [
  {
   "id": "biohub-esm--embed-protein-sequences-with-esmc",
   "title": "Turn protein sequences into embeddings with ESMC",
   "best": true,
   "url": "https://dedupedai.com/use-cases/biohub-esm/embed-protein-sequences-with-esmc/",
   "json": "https://dedupedai.com/use-cases/u/biohub-esm/embed-protein-sequences-with-esmc.json"
  },
  {
   "id": "biohub-esm--predict-protein-structures-with-esmfold2",
   "title": "Predict an all atom structure from a sequence",
   "best": false,
   "url": "https://dedupedai.com/use-cases/biohub-esm/predict-protein-structures-with-esmfold2/",
   "json": "https://dedupedai.com/use-cases/u/biohub-esm/predict-protein-structures-with-esmfold2.json"
  },
  {
   "id": "biohub-esm--interpret-esmc-features-with-sparse-autoencoders",
   "title": "Read which learned features a protein activates",
   "best": false,
   "url": "https://dedupedai.com/use-cases/biohub-esm/interpret-esmc-features-with-sparse-autoencoders/",
   "json": "https://dedupedai.com/use-cases/u/biohub-esm/interpret-esmc-features-with-sparse-autoencoders.json"
  }
 ],
 "disclaimer": "Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk."
}