Mithrl, the AI infrastructure company for biopharma, today announced a $20 million Series A led by Obvious Ventures, with participation from Headline, AGI House, and several pharma executives. The round funds Mithrl’s second-generation platform layered on its proprietary biomedical world model and paired with an agentic harness for model routing, token optimization, and context orchestration. Together, these AI infrastructure technologies called, Mithrl-1, let biopharma teams put validated biomedical context and best-in-class AI to work inside their own environment.

The results are already tangible. Discoveries powered by Mithrl have contributed to more than half a dozen customer-owned patent filings. Its biomedical world model delivers 16x more primary evidence per answer than frontier models alone and scored 0.96 for scientific correctness in expert-rated biomedical benchmarks, versus 0.60 without Mithrl’s proprietary knowledge system. Already deployed across top-10 pharma, clinical-stage biotechs, and genomics platform partners, Mithrl turns scientific insights into defensible intellectual property and gives teams greater confidence in which programs to advance.

Mithrl’s biomedical world model provides a knowledge system of biology that its agentic harness reasons over, curated and validated from peer-reviewed science, rather than extrapolated from public data alone. Agents work within the bounds of peer-reviewed literature, public and partnered datasets, and each client’s proprietary data, cross-referencing across disciplines to surface non-obvious connections and never departing from what has been published and verified. Every hypothesis comes with receipts and a confidence score: exactly what the system read, how it got there, and how much to trust it.

“Every player in this category is racing to generate more raw hypotheses, faster. But the industry is cracking under the weight of hypotheses it can’t triage or validate,” says Vivek Adarsh, Co-Founder and CEO of Mithrl. “We built Mithrl to answer a different question: which of these will actually hold up in downstream experiments, and ultimately in patients? The platform doesn’t guess. It reasons from trusted studies and each client’s in-house evidence the way a client’s best scientists do, and it shows its work every time. That’s why our enterprise clients trust it with program-level decisions, not just ideation.”

A world model is only useful when it is deeply integrated into the real workflows of discovery teams, so Mithrl doesn’t just ship its world model as an app. Mithrl-1 is forward-deployed directly into each client’s environment and harmonizes across the frontier models each team already trusts, never forcing model lock-in. Each client’s instance is extended with their own proprietary data and pipelines. No two biopharma organizations have the same requirements, and none should have to accept someone else’s defaults.

From there, client teams build and manage their own biomedical AI agents in-house, across modalities, without waiting on a vendor roadmap. Because those agents reason over validated biology instead of searching their way to an answer, they do more with far less compute. In a recent benchmark, Mithrl’s platform, running on its proprietary biomedical world model and harness, used 45% fewer tokens than standard workflows using the same frontier base models out of the box.

“The AI-for-biology stack is getting crowded at the app layer, but the durable advantage lives in the middle, the proprietary data, world models, and custom infrastructure each biopharma team needs,” says Rohan Ganesh, Partner at Obvious Ventures. “Everyone’s requirements are different. There is no one-size-fits-all, and that’s exactly where Mithrl is building.”

Mithrl-1 is available for early access now; visit https://pages.mithrl.com/mithrl-1.

About Mithrl

Mithrl is the AI infrastructure company for biopharma. Mithrl designs and deploys custom AI agents across the drug development pipeline and gives teams the ability to build and run their own agents in-house, all grounded in a proprietary biomedical world model built for reproducibility, auditability, and transparency. For more information, please visit www.mithrl.com or connect at, hello@mithrl.com.

Media gallery

About The Author
Business Wire