ChemistryGPU15 credits

DiffSBDD — 3D Structure-Based Drug Design

3D equivariant diffusion for pocket-conditioned molecule generation

DiffSBDD is an equivariant diffusion model that generates 3D molecules directly inside a protein pocket. Unlike SMILES-based generators (REINVENT), DiffSBDD produces geometry-aware candidates with realistic bond angles, distances, and conformations.

$0.15
per API call
15
credits per call
/v1/generate/molecules/diffsbdd
API endpoint

Features

3D pocket-conditioned generation
Equivariant diffusion (SE(3)-aware)
Realistic geometries out of the box
Complements Boltz-2 / Chai-1 for validation loop

Quick Start

DiffSBDD — Python Examplepython
# DiffSBDD endpoint coming in Sprint 52B.
# In the meantime, use /v1/generate/molecules (REINVENT4).

Use Cases

1

Structure-based drug design

2

Pocket-aware lead generation

3

Novel molecule scaffolds

Learn More

DiffSBDD: 3D Pocket-Conditioned Molecule Generation Explained

12 min read

Related Tools

REINVENT4

AI-powered de novo drug design and lead optimization

$0.08/call

Chai-1

Predict how drugs bind to protein targets at atomic resolution

$0.15/call

Boltz-2

Predict biomolecular complex structures: protein, ligand, DNA, RNA

$0.15/call

Start Using DiffSBDD

500 free credits every month. No credit card required.

Back to all tools