# chaidiscovery/chai-lab

Chai-1, SOTA model for biomolecular structure prediction

Repository: https://github.com/chaidiscovery/chai-lab
Canonical: https://ross.abutalabs.com/products/chai-lab
Homepage: https://www.chaidiscovery.com
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-06-30T07:48:29+00:00

## Health v2 (maintenance only)
Score: 65/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 90, release rhythm 40, longevity 51
- inputs: {"age_days": 723, "days_push": 64, "days_rel": 534, "gap_med": 5, "n_releases_24m": 16}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1986, forks 286 (observed 2026-08-28T04:06:02.769749+00:00)

## What it is
Chai-1 is a state-of-the-art multi-modal foundation model for biomolecular structure prediction, handling proteins, small molecules, DNA, RNA, and glycosylations in unified complexes. It is distributed as a Python package (chai_lab) with both a CLI and Python API for running folding inference on GPU hardware.

## Use cases
- predict 3D structure of a protein complex from FASTA sequences
- fold protein-ligand complexes including small molecules and nucleic acids
- run antibody structure prediction for drug discovery
- predict biomolecular interactions with MSAs and templates
- generate PDB structure predictions via Python API

## When to choose
- you need SOTA accuracy for multi-component biomolecular complexes
- you have access to a CUDA GPU with bfloat16 support (A100/H100/L40S class)
- you want both CLI and programmatic Python inference
- you work in drug discovery or computational biology

## When to avoid
- you only have CPU-only or non-Linux environments
- you need lightweight or fast structure prediction without heavy GPU resources
- you need a web service rather than a local inference package
- your focus is general-purpose protein modeling beyond structure prediction

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, llm-inference, cli
- domain: bioinformatics, machine-learning, artificial-intelligence, healthcare
- platform: python, cli
- tags: structure-prediction, protein-folding, drug-discovery, molecular-modeling, alphafold-alternative, cuda, linux, gpu

## Member repositories
- chaidiscovery/chai-lab (main) score 65

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:02.769749+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:03:06.206207+00:00, confidence not recorded.
  - readme: https://github.com/chaidiscovery/chai-lab (fetched 2026-08-28T04:06:02.769749+00:00, sha f064f1e9abfb)
  - homepage: https://www.chaidiscovery.com (fetched 2026-08-29T10:42:57.042689+00:00, sha 5d8bb39e8fd7)
- Data as of 2026-08-30T08:39:29.467469+00:00.