Protein Models
Protein language models and structure predictors have fundamentally changed structural biology by learning the evolutionary and physicochemical constraints encoded in amino acid sequences. From predicting three-dimensional folds with atomic accuracy to generating novel sequences with desired properties, these models now underpin drug discovery, enzyme engineering, and vaccine design pipelines. The field spans sequence-only language models like ESM to physics-informed structure predictors like AlphaFold, offering tools for nearly every stage of protein research.
83 models in this category
Notable Models
Top-rated protein models from our evaluations
DecoderTCR
Chan Zuckerberg Initiative
A masked language model for T-cell receptor and peptide-MHC interaction prediction using compositional pretraining and entropy-guided non-autoregressive decoding.
Boltz-2
MIT CSAIL / Recursion Pharmaceuticals
Open model that jointly predicts biomolecular structure and small-molecule binding affinity, approaching FEP+ accuracy in seconds on a single GPU.
Evolla
Westlake University
An 80B-parameter multimodal protein-language model that decodes protein function through natural language dialogue, integrating sequence, structure, and evolutionary context.
LigandMPNN
Institute for Protein Design
Protein sequence design method that explicitly models small molecules, nucleotides, and metals at atomic resolution, enabling ligand-aware design with 100+ validated designs.
Pinal
Westlake University
A 16B-parameter framework for de novo protein design from natural language, converting text descriptions into functional protein sequences via two-stage structure-conditioned generation.
ProGen3
Profluent
Sparse mixture-of-experts autoregressive protein language model family pretrained on 1.5 trillion amino acid tokens with compute-optimal scaling.