Chai Discovery

What Chai Discovery does

Chai Discovery builds the computer-aided design suite for molecules. They train frontier models that learn the underlying foundations of biochemical structure and interaction, enabling scientists to design new medicines on the computer. They are building a future where new medicines can be designed with the precision and speed of modern engineering.

Products

• Chai-2 series of modelsA series of models for de novo protein design that works reliably, enabling the design of new medicines.

Key capabilities

• De novo protein design that works reliably
• Specify antibody format (mAbs, VH-VL / VHH, etc.)
• Achieve high success rates (>10% for antibodies, >50% for miniproteins)
• Faster experimentation cycles (characterization in <2 weeks)
• Prompt with any target (public/private structures, predicted structures)
• Target specific antigen confirmations (agonism or antagonism)
• Address challenging targets (membrane proteins)
• Design for epitope specificity
• Handle chemical modifications (bound ligands, PTMs)
• Specify advanced design criteria (species/ortholog specificity, cross-reactivity)

Use cases

• AI-driven antibody discovery
• Accelerate drug discovery with AI
• Advance AI-driven immunology breakthroughs
• Drug-like antibody design against challenging targets with atomic precision
• Designing new medicines on the computer
• Bypass screening in protein design
• Targeting small, buried, or non-immunogenic epitopes

AI approach

Chai Discovery builds computer-aided design suites for molecules, training frontier models that learn the underlying foundations of biochemical structure and interaction. This enables scientists to design new medicines on the computer, specifically focusing on de novo protein design, including antibodies and miniproteins, with high success rates and faster experimentation cycles.

Tech named: Chai-2 series of models

Industries served

• Pharmaceutical
• Biotechnology
• Life Sciences

What it says sets it apart

• Frontier models learning biochemical structure and interaction
• High success rates for de novo protein design
• Faster experimentation cycles compared to traditional methods
• Ability to target specific antigen confirmations and challenging targets
• Precision and speed in medicine design

Funding rounds we track

From the AI funding tracker — rounds as reported by the linked publications.