All Competitors

Protein-ligand binding affinity prediction from sequence and SMILES, without MSAs. Coarse-grained cofolding runs over 10x faster than Boltz-2.

Protein language model pretrained on structural domain segments, encoding fold and contact signals for remote-homology detection from sequence alone.

All-atom foundation model for macrocyclic peptide structure prediction, permeability estimation, and de novo design across non-canonical chemistries.

Variable-length generative protein design across structure, sequence, motif scaffolding, and peptide co-design via a generalized Poisson flow.

Generative language model that designs drug-like SMILES conditioned on disease ontology and a target protein sequence for de novo drug discovery.

All-atom foundation model for immune-receptor design that predicts structures and co-designs CDR sequences for antibodies, nanobodies, and TCRs.

Open-source all-atom co-folding foundation model for protein-ligand, protein-protein, and antibody-antigen complex prediction in drug discovery.

T-cell receptor-MHC restriction prediction from amino acid sequence, mapping TCRs to their restricting HLA allele at 0.97 held-out AUC.

Structure-based drug design language model fusing protein structural and evolutionary encoders with SAFE fragment tokens for hit-to-lead generation.

Heterograph foundation model for metabolic engineering, fusing protein, chemistry, and text embeddings over a 6.9M-node biological knowledge graph.

Atomic protein model building from cryo-EM density maps, resolving conformational heterogeneity through atom-centric sampling and diffusion.

Diffusion model for 3D small-molecule design against protein-protein interaction sites, guided by the natural binding peptide or protein partner.

Text-guided localization model that grounds natural-language functional descriptions to specific residue regions of a protein sequence.

Multimodal molecular generation model for drug design, conditioned on properties, pharmacophores, protein sequences, or protein binding pockets.

Diffusion model that generates 3D small molecules conditioned on protein pockets and partial fragments encoded as continuous spatial density maps.

Decoder-only foundation model that unifies sequences, 3D structures, and natural language for small molecules and proteins in one shared token space.

Small-molecule hit-discovery pipeline using Boltz-2 co-folding and affinity prediction to rank in-stock compounds or make-on-demand chemical space.

De novo protein binder and nanobody design pipeline that ranks candidates by a protein-protein interaction model rather than structural confidence.

Gene representation framework fusing DNA, transcript, protein, text, and single-cell embeddings into one latent space that survives missing views.

Conditional denoising diffusion model that designs antigen-specific TCR CDR3β sequences conditioned on peptide-MHC targets and germline V-genes.

Generative transformer for phylogenetic inference that transduces sets of unaligned molecular sequences directly into Newick-format trees.

Protein binder generator producing receptor-conditioned binders from sequence alone, using a sparse Mixture-of-Experts transformer with no 3D input.

Reinforcement learning framework that fine-tunes the ProGen2-OAS antibody language model with GRPO to cut germline bias in generated sequences.

Message-passing neural network that designs buried hydrogen-bond networks onto protein backbones, combining learned placement with PyRosetta scoring.