TheTechno Optimist.

Drug design engine and pipelineIsomorphic Labs—Isomorphic Labs in Fundraising Discussions at $40–50 Billion ValuationDrug design engine and pipelineIsomorphic Labs—Isomorphic Labs describes its drug design engine, reporting double AlphaFold 3's accuracy on a protein-ligand benchmarkIsomorphic Labs—Isomorphic Labs reportedly raises $2.1B Series B

Isomorphic Labs

AI drug design, AlphaFold 3

GBAlphabet-ownedest. 2021isomorphiclabs.com (opens isomorphiclabs.com)Checked 25 Sep

The brief

Analyst view
  • Raised its first outside money, $600M led by Thrive Capital, in Mar 2025, then a reported $2.1B in May 2026.
  • The goal of AI-designed drugs in trials by end of 2025 slipped; now expected by end of 2026.
  • Its Lilly and Novartis deals (Jan 2024) could pay up to about $3B if goals are hit.

Technical approach

As reported
Structure

Predicting how molecules fit

AlphaFold 3 predicts the 3D shape of a protein with a drug, DNA or RNA stuck to it, so chemists can see how a candidate might fit.

Affinity

How tightly it grips

Fitting isn't enough; more models estimate how strongly a molecule binds and whether it behaves like a drug. That's still the hard part.

Design loop

Generate, predict, test

AI proposes molecules, other models rank them, lab tests feed back to improve the AI, aiming for fewer rounds to reach a drug candidate.

Programs

Isomorphic Labs, two programs

Updated 2d ago
Drug design engine and pipelineA set of AI models that predict protein shapes, binding and drug properties, used to design new drugs for its own pipeline and for partners.ResearchNextFirst AI-designed drug in clinical trials · Dec 2026+12 moUpdated 2d agoView program →AlphaFold 3An AI that predicts the 3D shapes of proteins, DNA, RNA and small molecules fitting together. Built by Google DeepMind and Isomorphic Labs.ScalingUpdated 9 Oct 2024View program →
Research · Proto · Pilot · Deploy · Scaling

Research & demos

3 items

Open technical problems

Open · Active · Partial
  • Predicting what works in peopleOpen

    Most drugs fail because they don't work in humans or are toxic, which today's models predict poorly.

  • New targets and new chemistryActive research

    The models are best on things like their training data; unusual proteins and new kinds of molecules are still hard.

Peers

AI drug discovery · 3

Company & funding

2 rounds · as reported
Raised
$2.7Bdisclosed private rounds
Last round
Series BMay 2026
Parent
Alphabet
HQ
GB
Founded
2021
DateRoundAmountPost-moneyLead
May 2026Series B$2.1B——
Mar 2025First external round$600M—Thrive Capital

Partners & customers

1 partner · 2 customers · 1 investor

People

Key people
Primer