# Generate Biomedicines

**Canonical:** https://apis.io/providers/generate-biomedicines/  
**Website:** https://generatebiomedicines.com  
**APIs profiled:** 0

Generate Biomedicines is a Somerville, Massachusetts therapeutics company pioneering Generative Biology, a machine-learning-driven approach to de novo protein design across antibodies, peptides, enzymes, and gene therapies. Founded out of Flagship Pioneering and chaired by Noubar Afeyan, the company trains generative models on the full compendium of natural protein structures and sequences plus proprietary wet-lab data to design novel protein therapeutics computationally rather than by trial-and-error discovery. Its open-source flagship is Chroma, a programmable diffusion-based generative model for joint protein structure and sequence design published in Nature 2023 ("Illuminating protein space with a programmable generative model", doi:10.1038/s41586-023-06728-8) and released on GitHub under Apache 2.0 with pretrained weights available under a custom Chroma Parameters License (free for academic and nonprofit use). Chroma uses equivariant graph neural networks, diffusion modeling, and conditional random fields to sample all-atom protein structures with sub-quadratic complexity, supporting conditional generation via composable Conditioner modules for symmetry, substructure, shape, and secondary-structure constraints. The internal Generate Platform iterates on a "generate, build, measure, learn" loop that pairs in-silico design with high-throughput protein synthesis and characterization; the company reports having generated, built, and tested more than 42,000 proteins. Generate Biomedicines has platform collaborations with Amgen (six programs, January 2022) and Novartis (multiple programs, September 2024) and a CAR T collaboration with Roswell Park. Its clinical pipeline is led by GB-0895, a long-acting anti-TSLP antibody for severe asthma now in Phase 3 (SOLAIRIA study) with a parallel Phase 1 in COPD, plus preclinical programs GB-4362 (anti-free-MMAE for ADC oncology safety) and GB-5267 (MUC16 armored CAR T for ovarian cancer). The company priced its initial public offering in February 2026. Revenue derives from drug development, partner collaborations, and (for Chroma) commercial weights licensing; there is no public developer REST/HTTPS API or SDK for the Generate Platform itself — Chroma is consumed as a Python library installed via pip with model weights downloaded from chroma-weights.generatebiomedicines.com after registering for an API key.

## Kin Score — 9.1 / 100 (minimal)

Scored 2026-08-20 under rubric 0.12.0. Trend: flat (+0.0 from 9.1).

| Facet | Score |
|---|---|
| Discoverability | 50.0 |
| Contract Quality | 0.0 |
| Governance | 0.0 |
| Contract Governance | 0.0 |
| Operational Transparency | 2.6 |
| Developer Ergonomics | 19.0 |
| Commercial Clarity | 0.0 |
| Access Clarity | 0.0 |

## Agent readiness — 0.0 (human-only)

| Dimension | Value |
|---|---|
| Spec Presence | no |
| Agentic Access | no |
| Reversibility Documented | no |
| MCP Server | no |
| Auth Clarity | no |
| Idempotency | no |
| Error Semantics | no |
| OpenAPI Examples | no |
| Rate Limit Signal | no |
| Event Surface Described | no |
| Agent Skills | no |
| Well Known Catalog | no |
| Consent Identity | no |
| Agent Card | no |
| Dry Run Mode | no |

## Access

Unknown — onboarding: unknown, pricing: unknown, trial: no (confidence: low).

## Security (1)

- **Generate Biomedicines Domain Security** — TLSv1.3 · DMARC

## Tags

Generative Biology, Protein Design, De Novo Design, Diffusion Models, Antibodies, Drug Discovery, Therapeutics, Machine-Learning, Biotechnology, Chroma, Foundation Models, Equivariant Neural Networks, Flagship Pioneering

---

Profiled by [API Evangelist](https://apievangelist.com) and published on [APIs.io](https://apis.io/providers/generate-biomedicines/). Scores are computed from the provider's own public artifacts under a published rubric.