Aureka Biotechnologies Raises $100M Series B for Biological AI

Aureka Biotechnologies closed a $100M Series B on August 10, 2026, to expand the biological foundation models and experimental systems behind its AI-driven drug-discovery platform. Granite Asia funded the first tranche exclusively, while an unnamed strategic investor led a later tranche with participation from HighLight Capital and follow-on investment from MPCi and NRL Capital.

The round brings Aureka's company-reported total funding to nearly $200M since its 2023 founding. The money is aimed at a thesis that sounds simple until biology gets involved: a drug-discovery model becomes more useful when it can learn from the physical experiments that reveal whether its designs actually function.

What Aureka Raised and Who Backed It

The Series B was completed in multiple tranches. Granite Asia exclusively funded the first, an unnamed strategic investor led the next, and HighLight Capital joined alongside returning shareholders MPCi and NRL Capital. Aureka did not disclose its valuation or the identity of the strategic investor, which means the cleanest reading of the deal is also the least cinematic: $100M of fresh capital, a substantial cross-border syndicate, and several important terms that remain private.

Aureka says the financing will go primarily toward research and large-scale training for the next generation of its biological foundation models. The stated technical priorities include de novo molecular design, biological structure modeling, and function prediction, plus an expansion of the company's Lab-in-the-Loop system. That allocation matters because the expensive part of AI drug discovery is not merely generating a plausible molecular answer; it is producing enough high-quality experimental evidence to know when the answer survives contact with biology.

Why the Wet Lab Is Part of the AI Stack

Aureka's strategy joins models, AI agents, digital biology, proprietary single-cell screening, high-throughput validation, and drug-development workflows in one feedback system. Models propose molecules and scientific hypotheses, experiments test those ideas, and the resulting functional data returns to model training and project-specific optimization. In that design, the laboratory is not a cleanup crew waiting after inference. It is part of the learning architecture.

That distinction is the core business argument behind the round. Public biological datasets can help many companies train capable models, but proprietary experimental feedback can create a data asset that competitors cannot download from the same repository. If Aureka can repeatedly turn model output into useful experiments and those experiments into better future designs, the company is building a compounding system rather than a collection of disconnected prediction tools.

OpenDDE Makes Part of the Thesis Inspectable

Aureka's proprietary foundation model is called AuraIDE, while OpenDDE is its open-source counterpart. The repository describes OpenDDE as an all-atom biomolecular foundation model for structure prediction, design, and optimization, but it also states that the software is a preview whose interfaces and checkpoints may change. That warning belongs in the same paragraph as the ambition because a public model is evidence of engineering progress, not proof that a drug candidate will work in people.

Independent testing provides a stronger technical signal than company marketing alone. Tamarind Bio reported that OpenDDE v1 reached 76.1% DockQ success on FoldBench v1, compared with 47.9% for previously published AlphaFold 3 results on identical inputs and setup. The benchmark covers antibody-antigen structure prediction, so it can support the claim that OpenDDE is technically competitive on that task; it cannot establish clinical safety, efficacy, manufacturability, or regulatory success.

Weian Zhao's Translation Problem

Founder and CEO Weian Zhao approaches the company from both academic science and commercial translation. UC Irvine lists Zhao as a professor of pharmaceutical sciences whose work includes diagnostics, biosensors, microtechnology, and cell therapies, with chemistry degrees from Shandong University and McMaster University. Aureka's origin is tied to the gap between promising scientific results and medicines that can reach patients, a gap that turns every elegant model into a long operational argument.

The Series B gives Zhao's team more room to pursue what the company calls a biological world model. In Aureka's roadmap, that means systems that could reason about interactions among molecules, predict consequences of biological interventions, and direct agents through repeated design and testing cycles. The phrase should be treated as a direction rather than a finished capability; today's verifiable work is narrower and centers on biomolecular structure prediction, antibody design, and experimental feedback.

Commercial Evidence, With the Fine Print Intact

Aureka reports strategic partnerships with multiple global pharmaceutical companies and tens of millions of dollars in revenue over the past two years. Those claims suggest the platform is touching commercial programs instead of existing only inside benchmark leaderboards, but the announcement does not name customers, break down contracts, or provide audited revenue. The signal is meaningful, while the visibility remains limited.

That balance is common in platform biotechnology. A company can have sophisticated models, partner interest, and encouraging preclinical work while still facing long timelines through replication, animal studies, manufacturing, clinical trials, and regulation. Aureka's new capital can accelerate model training and experimental throughput, but it does not repeal the slowest parts of therapeutics development.

What the $100M Series B Signals

The broader market signal is that AI drug discovery is becoming an infrastructure contest. Standalone models matter, but investors are increasingly looking for the systems around them: proprietary data generation, automated experiments, feedback loops, and a path from computation to developable assets. Aureka's financing places a large bet on the idea that the wet lab can be as strategically important as the model because the lab produces the ground truth that keeps the model honest.

The opportunity is real, and so is the test. Aureka now has capital, an inspectable open-source model, a founder with deep translational experience, and company-reported commercial traction. What comes next will be judged less by the size of the round than by whether the closed loop produces molecules that remain compelling after the benchmark ends, the experiments multiply, and biology starts charging full price for every assumption.