Download
OpenFold carefully reproduces (almost) all of the features of the original open source inference code (v2.0.1). The sole exception is model ensembling, which fared poorly in DeepMind's own ablation testing and is being phased out in future DeepMind experiments. It is omitted here for the sake of reducing clutter. In cases where the Nature paper differs from the source, we always defer to the latter. OpenFold is trainable in full precision, half precision, or bfloat16 with or without DeepSpeed, and we've trained it from scratch, matching the performance of the original. We've publicly released model weights and our training data — some 400,000 MSAs and PDB70 template hit files — under a permissive license. Model weights are available via scripts in this repository while the MSAs are hosted by the Registry of Open Data on AWS (RODA).
Features
- Faster inference on GPU, sometimes by as much as 2x
- Inference on extremely long chains, made possible by our implementation of low-memory attention
- Custom CUDA attention kernels modified from FastFold's kernels support in-place attention during inference and training
- Efficient alignment scripts using the original AlphaFold HHblits/JackHMMER pipeline or ColabFold's
- FlashAttention support greatly speeds up MSA attention
- OpenFold also supports inference using AlphaFold's official parameters
Project Samples
