Protein Design · Meta FAIR
Language-model-informed inverse folding — an alternative view to ProteinMPNN.
1–30s
Runs on On-demand GPU
Runs on our servers with on-demand compute. A first run needs time to load the model; active capacity can be reused and scales down when idle.Run this model
About this model
ESM-IF1 designs sequences for a backbone using ESM2 representations rather than a graph network. Running it alongside ProteinMPNN and taking the consensus is a cheap way to raise design confidence before committing to synthesis.
Standardized I/O contract
Every model in the Hub speaks the same contract, which is what lets the Router and the agent call any of them without special-casing.
Inputs
Backbone (PDB)
Chain
Temperature
Sequences
Outputs
Sequences with model perplexity and recovery.
