Overview
<cite index="4-6">Chai Discovery, the company engineering artificial intelligence (AI) models to discover new molecules, announced a $400 million Series C fundraise to further accelerate progress.</cite> <cite index="7-1">The raise values the AI drug discovery company at $3.8 billion — nearly triple the $1.3 billion valuation it held seven months ago.</cite>
Investors and Capital Structure
<cite index="1-8">The round was led by Index Ventures, with participation from Kleiner Perkins, Sequoia Capital, and Dimension, alongside new investors including Bain Capital Ventures, Battery Ventures, Baillie Gifford, BDT & MSD, Sapphire Ventures, and Avra Capital.</cite> <cite index="1-9">Existing backers Thrive Capital, OpenAI, Oak HC/FT, Menlo Ventures, General Catalyst, Glade Brook, Avenir, Lachy Groom, and Yosemite also participated.</cite> <cite index="1-11">Chai Discovery has raised approximately $630 million across four rounds since its founding in 2024.</cite>
<cite index="1-10">The company said proceeds will be used to further accelerate its AI model development and expand its platform capabilities, though specific allocation details were not disclosed.</cite>
Funding History and Founding
<cite index="2-15">Chai Discovery was founded in 2024 by Joshua Meier, Jack Dent, Matthew McPartlon, and Jacques Boitreaud.</cite> <cite index="2-16">The company was established by founders with backgrounds spanning OpenAI, Meta FAIR, Stripe, molecular design, and academic research.</cite> <cite index="1-1,1-2">A seed round of approximately $30 million, backed by Thrive Capital, OpenAI, and Dimension, was followed by a $70 million Series A in August 2025 led by Menlo Ventures. A $130 million Series B co-led by Oak HC/FT and General Catalyst followed in December 2025, bringing the company to unicorn status at a $1.3 billion valuation.</cite>
Technology Platform
<cite index="1-12">The platform's foundation model (FM), Chai-1, performs multimodal biomolecular structure prediction across proteins, small molecules, DNA, and RNA simultaneously, including their interactions.</cite> <cite index="1-13">Chai-2, released in 2025, added a generative layer: a full-atom diffusion architecture capable of designing complete antibody sequences and structures from scratch, conditioned only on a target protein and a specified binding epitope.</cite> <cite index="1-14">The company reported that Chai-2 achieved experimental hit rates of approximately 16–20% in fully de novo antibody design — compared to rates below 1% for prior computational methods — a threshold that makes small-panel experimental validation practical and commercially viable.</cite> <cite index="1-15">Chai-3, the current generation, is reported to reduce the antibody design failure rate by approximately 50% relative to Chai-2 and to improve binding affinity and multi-specific molecule engineering.</cite>
Commercial Partnerships
<cite index="1-5">In January 2026, the company announced a research collaboration with Eli Lilly and Co. that included development of a bespoke AI model trained on Lilly's proprietary data.</cite> <cite index="1-6">A license agreement with Pfizer, announced in June 2026, granted Pfizer early access to Chai-3, the company's most advanced model, alongside a custom model tailored to Pfizer's workflows.</cite> <cite index="1-7">The day before the Series C was announced, Chai disclosed a collaboration with Novartis to advance AI-driven antibody discovery, extending its roster of major pharma partners to three.</cite>
Market Context
<cite index="7-7">The global AI drug discovery market is expected to reach $2.35 billion in 2025 and grow to $13.7 billion by 2033, at a compound annual growth rate (CAGR) of roughly 25%.</cite> <cite index="11-8">Isomorphic Labs, Google DeepMind's drug discovery spinout, closed $2.1 billion in May and also counts Novartis as a partner, meaning the pharma giant is now backing two competing approaches to AI-designed medicine at once.</cite> <cite index="11-18,11-19">Drug discovery remains one of science's highest-risk industries, where promising molecules must still survive years of laboratory validation, clinical trials, and regulatory scrutiny before reaching patients — a timeline that does not compress simply because a model can generate candidates faster.</cite>