Knit Health raises $35M to expand Med-BEAM patient care AI model

What happened
Knit Health announced on November 16, 2026, that it raised $35 million in Series A funding led by NEA. The funding will be used to expand its Med-BEAM healthcare world model and deploy it in more clinical settings. Med-BEAM is designed to predict patient care pathways and support clinical operations. The company benchmarked Med-BEAM against Anthropic's Claude models, using real hospital admission records.
Key details
- Med-BEAM achieved an average precision score of 0.565 in predicting 24-hour patient actions.
- Med-BEAM's estimated processing time per patient was about two seconds, compared to 216–408 seconds for the Claude models.
- Estimated per-patient cost for Med-BEAM was 37 to 257 times lower than the Claude models.
- The Series A round follows an $11.6 million seed round and a pre-seed round led by Moxxie Ventures.
Why it matters
Healthcare providers, technology companies, and developers could use Med-BEAM as a shared foundation for clinical decision-making tools, potentially reducing operational costs and improving efficiency in patient care coordination. The reported accuracy, speed, and lower cost of Med-BEAM compared to leading language models may allow hospitals and clinics to automate and streamline processes like patient routing and scheduling. This funding enables Knit Health to scale its model to more clinical environments, affecting how care pathways are managed across different institutions.

