AI × Bio

AI Models for Scientists

Use our model API and fine-tuning services to access infinitely scalable compute on the latest models. No commercial use restrictions. No data retention or IP requests.

Get Started

Cutting-Edge AI × Bio Research

01

Antibodies

Our protein engineering team has an extensive focus on antibody machine learning models, including generative models for antibody design and predictive models for antibody binding and developability properties.

02

Genetic Medicines

We're laser focused on building new foundation models with applications in mRNA design and broadly in genetic engineering. Our first model allows for the generative design of 3’ UTRs.

03

Enzyme Function

Ginkgo has an extensive history in enzyme optimization, and we are working to enable new possibilities in enzyme design with our models.

Model API

Introducing Ginkgo’s Model API

A powerful tool designed to bring biological AI models directly to your AI team. Our API is programmer-friendly and ultra-low cost. We’re making both our internally-developed AI tools and open source models available to everyone. The API provides an easy and scalable way to access sophisticated models trained on protein and DNA data.

ai model api

AI × Biology

Explore Ginkgo's AI Products

Model API
Learn More
Fine-tuning Services
Learn More
Compute Services
Learn More

Model

Ginkgo’s First Public Model: AminoAcid-0

Our first protein sequence LLM — AA-0 — is a 650M parameter model following the ESM-2 architecture, trained on public data and 2 billion proprietary sequences from Ginkgo’s genomic database.

Partnership

Our Unique Partnership with Google

Ginkgo’s unique multi-year partnership with Google gives us access to capacity reservations in the latest in the GPU and TPU technology. We’re make our cloud infrastructure directly available to our customers. Partner with us to accelerate your strategic cloud goals for model training in biology.

We’re building a future where access to models is easy and scientists can focus on using models to solve scientific problems. We’d love to hear from you on how to make this possible.

Build with Us
Chat with Our Team