ByteDance

All-atom protein co-design model that generates sequence and structure together in one unified diffusion process, aimed at hard binder design.

464M-parameter structure prediction and design model that improves antibody-antigen complex accuracy over Protenix-v1 and adds generative VHH design.

All-atom generative foundation model that designs small molecules, peptides, and nanobodies against a target binding site from a single checkpoint.

De novo protein binder design suite from ByteDance pairing diffusion and hallucination generators with confidence-based filtering of designs.

Simplex diffusion model for discrete sequence generation, with released checkpoints for DNA enhancer design and de novo protein sequence design.

Structure-based drug design framework pairing pharmacophore-guided latent diffusion with training-free, pocket-aware evolutionary optimization.

Cyclic peptide design conditioned on target protein structure, generating all four cyclization types via all-atom, all-bond harmonic SDE modeling.

Protein conformation and dynamics generation from MD data, sampling trajectories, independent ensembles, and interpolations between two known states.

All-atom generative model for protein complexes that designs multi-chain binders from scratch and performs multimer folding and inverse folding.

Multimodal diffusion protein language model co-generating sequence and structure. Bit-level structure supervision cuts folding RMSD from 5.52 to 2.36.

De novo binder design across small molecules, peptides, and antibodies from one geometric latent diffusion model over graphs of molecular blocks.

Open-source PyTorch reproduction of AlphaFold 3 under Apache 2.0, matching or exceeding AF3 on protein-ligand, protein-protein, and RNA benchmarks.

Target-conditioned peptide binder design model that samples hot-spot residues from an energy-based density, then extends fragments autoregressively.

Two-stage backbone generator that designs protein domains separately, then weaves them into one long chain with an SE(3) diffusion assembly module.

Multimodal discrete-diffusion protein language model that co-generates amino acid sequence and 3D backbone structure from a single transformer.

Generative foundation model for cryo-EM density maps using flow matching, enabling zero-shot denoising, map sharpening, and missing wedge restoration.

Structure-based drug design diffusion model that re-extracts the essential binding subcomplex from a pocket at every step of 3D ligand generation.