Latent Labs has launched its artificial intelligence-powered drug design platform to researchers worldwide, providing free daily access to a laboratory-validated system capable of generating therapeutic protein designs from natural language prompts.

The platform centres on Latent-Y, an AI drug design agent that enables researchers to generate antibodies, nanobodies and other protein-based therapeutics using written descriptions of therapeutic targets or scientific publications. Every approved researcher will receive a daily allocation of 250 free designs, while larger research programmes can purchase additional design credits.

Unlike many AI platform launches, Latent Labs has supported the release with laboratory validation and results from independent academic research groups. According to published findings, Latent-Y achieved a 67% target-level success rate across nine drug discovery targets, producing protein binders with single-digit nanomolar binding affinities while reducing design workflows from weeks to hours.

Protein design has become one of the fastest-growing applications of artificial intelligence in drug discovery. Rather than screening vast libraries of existing molecules, AI systems can generate entirely new proteins designed to interact with specific biological targets, potentially accelerating the development of new medicines.

Before the public launch, the platform was evaluated by research groups working in structural biology, cancer immunotherapy and infectious diseases.

At the University of California, Davis, researchers designed nanobodies targeting a human ion channel. Four of the five candidates tested in the laboratory successfully blocked electrical current, while the strongest demonstrated an IC50 value of 30 nM, a level of potency that often requires multiple rounds of laboratory optimisation.

Researchers at LMU University Hospital incorporated AI-designed binders into CAR-T cells that successfully killed tumour cells in vitro. The programme has since progressed into animal studies.

Meanwhile, scientists at the Translational Genomics Research Institute validated dozens of high-affinity binders targeting a conserved malaria parasite protein that is considered difficult for the parasite to evade through mutation.

Simon Kohl, chief executive officer and founder of Latent Labs, said: “We built Latent-Y to be a force multiplier for researchers.”

He added that the platform enables researchers to run multiple design campaigns in parallel using expert-level biological reasoning while maintaining full control over each stage of the design process.

Latent Labs said the browser-based platform requires no specialist computational infrastructure, allowing researchers with limited computational biology experience to design therapeutic proteins using natural language. Researchers retain ownership of all sequences generated through the platform, and the company said user-generated data will not be used to train future AI models.

The launch reflects growing interest in applying artificial intelligence to early-stage drug discovery, where advances in protein design are enabling researchers to identify potential therapeutic candidates more quickly while reducing reliance on traditional laboratory screening approaches. By combining laboratory validation with broader researcher access, Latent Labs is aiming to make advanced computational drug design available to a wider scientific community.