Tamarind Bio
No-code and API inference for protein folding, antibody design, and docking — run ProteinMPNN, AlphaFold 2, RFdiffusion without setup.
Overview
Tamarind Bio is a no-code and API platform for running protein modeling tools without provisioning GPUs or managing containers. It targets researchers who want to run a protein structure prediction or protein design job the same way they would submit a form: pick a tool, supply a sequence or structure, and collect results. For teams that would rather call an endpoint, the same tools are exposed through a REST API, making Tamarind a fast route from an idea to a batch of predictions.
What you can run on Tamarind Bio
The catalog leans heavily into protein design and structure. ProteinMPNN performs inverse folding, generating sequences that fold to a target backbone, while RFdiffusion generates novel backbones for de novo design and motif scaffolding. Germinal supports antibody design workflows, and BoltzGen extends generative design toward binders and complexes. AlphaFold 2 covers protein structure prediction from sequence. Together these span the protein and antibody design loop — backbone generation, sequence design, folding, and validation — across both individual proteins and therapeutic candidates.
Running inference on Tamarind Bio
All of these run as hosted inference jobs: no local install, no weights to manage. You can drive them through a no-code web interface, ideal for wet-lab scientists and non-programmers, or through a REST API for scripted, high-throughput batch runs and integration into agentic pipelines. Every tool lives at a stable per-tool page so links stay durable. The platform is built for running models rather than distributing weights or hosting custom fine-tunes, which makes it a good fit for teams that need dependable turnkey access to a broad protein toolkit and want to compare several design and folding methods side by side.
Run inference on Tamarind Bio (5)
Protein structure prediction model that folds amino acid sequences into 3D structures with atomic accuracy, scoring a median GDT of 92.4 at CASP14.
Message passing neural network for fixed-backbone protein sequence design. Achieves 52.4% native sequence recovery, far surpassing Rosetta's 32.9%.
De novo protein design diffusion model that generates backbone structures conditioned on binding targets, symmetry constraints, and functional motifs.
All-atom generative model for de novo protein and peptide binder design against diverse biomolecular targets, wet-lab validated across 26 targets.
Generative pipeline for epitope-targeted de novo antibody (nanobody) CDR design that yields nanomolar binders from only dozens of designs per antigen.