Biology and chemistry

ESM (ESM-2, ESM3, ESM C, ESMFold2)

Protein language models: they understand amino acid sequences, predict structure (ESMFold2) and help with protein design. Since 2026 all open versions are under MIT.

Developer
EvolutionaryScale / Chan Zuckerberg Biohub (ESM-2 — Meta AI), USA
First release
Sep 2022
Latest release
May 2026
Sizes
8M – 15B (ESM-2), 300M – 6B (ESM C), 1.4B (open ESM3)
License
Commercial use allowedMIT (all published weights, including ESM3 open and ESM C; the large ESM3 98B is cloud-only)
Running
On your own serverAlso runs without a GPU
Industries
Healthcare, Science and research
Deploy on your server Hugging Face GitHub

What it does

  • Protein embeddings for predicting properties (stability, solubility)
  • Predicting 3D structures of proteins and complexes
  • Screening enzyme and antibody design candidates before lab work

Where it is used

Pharma and biotechIndustrial biotechnology (enzymes)Research institutes

Hardware requirements

LaptopLaptop or regular PC, up to 8 GB of VRAM — smaller versions
fits
1 GPUOne GPU with 16–80 GB — mid-size versions
fits
ClusterServer with several GPUs — flagship versions
no versions

Versions

  1. ESMFold2
  2. ESM C 6B
  3. ESM C 300M и 600M
  4. ESM3 open 1.4B
  5. ESMFold (Meta)
  6. ESM-2 (Meta)

How to run it

On your own serverWeights are downloaded from Hugging Face and served with vLLM (load and API) or llama.cpp (modest hardware). The model then runs in a closed network with no per-request fees.Hugging Face
How much hardware you needCalculate the VRAM for the model size, context length and number of concurrent requests.Open the hardware calculator

I can set this up end to end: pick the model size, deploy it on your server and connect it to your systems.

Frequently asked questions

Can ESM (ESM-2, ESM3, ESM C, ESMFold2) be used in a commercial project?

Yes. License: MIT (all published weights, including ESM3 open and ESM C; the large ESM3 98B is cloud-only). It allows commercial use, but it is still worth having a lawyer review the license before launch.

What hardware does ESM (ESM-2, ESM3, ESM C, ESMFold2) need?

At minimum: Laptop or regular PC, up to 8 GB of VRAM — smaller versions. Some versions also run on an ordinary CPU, without a GPU. You can calculate the exact VRAM for your model size and context in the hardware calculator.

Does ESM (ESM-2, ESM3, ESM C, ESMFold2) support Russian?

Language does not matter for this model: it does not work with text.

Where can I download ESM (ESM-2, ESM3, ESM C, ESMFold2) and what does it cost?

The ESM (ESM-2, ESM3, ESM C, ESMFold2) weights are open and free to download. You only pay for the hardware it runs on and for the setup. Source links are at the bottom of this page.

How I deploy it for clients

  1. SelectionI pick the model size for your task and hardware and test it on your examples.
  2. DeploymentI deploy it on your server or in a closed network and provide an API.
  3. Fine-tuningI fine-tune it on your data (LoRA) or connect a knowledge base — whichever is cheaper for the task.
  4. IntegrationI connect it to your CRM, ERP, bot, website or team chat and set up monitoring.

Similar models

Biology and chemistryBoltz (Boltz-1, Boltz-2, BoltzGen)MIT (Jameel Clinic) and Recursion · USACommercial use allowed

An open MIT-licensed alternative to AlphaFold 3: predicts structures of protein, DNA and small-molecule complexes; Boltz-2 estimates binding strength, BoltzGen designs new binding proteins.

DetailsBiology and chemistryOpenFold / OpenFold3AlQuraishi Lab (Columbia University) and the OpenFold consortium · USACommercial use allowed

A fully open reproduction of AlphaFold 2 and then AlphaFold 3 under Apache 2.0, with training data. OpenFold3 predicts complexes of proteins, nucleic acids and ligands.

DetailsBiology and chemistryArc Institute Evo / Evo 2Arc Institute (with Together AI, Stanford, NVIDIA) · USACommercial use allowed

DNA language models with context up to a million nucleotides: they assess the impact of mutations, annotate genomes and generate sequences. Evo 2 is trained on genomes from all domains of life.

Details

Source: github.com/Biohub/esm#licenses. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.