#146146th of 214 in ML frameworks and librariesFor science, health and research, science
colabfold
Protein structure prediction with AlphaFold2 and other models using MMseqs2 MSAs
Live demo ↗
(opens in a new tab)Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 3k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Sep 2026
- Language
- Jupyter Notebook
Missed the latest GitHub check; these facts may be older.
Overview
ColabFold packages AlphaFold2 and related folding models (OpenFold3, ESMFold, RoseTTAFold2, Boltz, BioEmu) as Google Colab notebooks and as a pip-installable colabfold_batch / colabfold_search CLI. MSAs come from the public MMseqs2 server or from locally built databases, and can be exported as AlphaFold3 JSON. Local runs use JAX, with optional fused kernels and a compilation cache.
Who it is for: Biologists and bioinformaticians predicting protein structures
Strengths
- Runs in Google Colab notebooks with no local setup
- MMseqs2 MSAs from the public server or local databases, with optional GPU search
- Fused-kernel option runs AlphaFold2 about 2.5x faster on Ampere or newer GPUs
- Exports MSAs as AlphaFold3 JSON, with DNA, RNA and ligands via FASTA syntax
Weaknesses
- Local MSA databases need about 940GB of disk and roughly 128GB RAM for searches
- Public MSA server allows only serial queries from a single IP
- Several notebooks are marked beta or in development
- Each new sequence length or MSA depth triggers a slow recompile
What it needs
- RAM ≥ 128 GB
- GPU optional
- Docker
- Needs MMseqs2 18 (8cc5c), JAX, OpenMM, conda, CUDA 12 or 13
- Models: AlphaFold2, AlphaFold2-multimer, OpenFold3, ESMFold, RoseTTAFold2
Also in ML frameworks and libraries
See all 214| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | ultralyticsPython library and CLI for training and running YOLO vision models | 94 out of 100 |
| 3 | pytorch-image-modelsPyTorch library of image models, pretrained weights and training scripts | 94 out of 100 |
| #4 | pytorchPython tensor library with GPU support and tape-based autograd | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 92 out of 100 |