Baseten surged to a $2.15 billion valuation with a $150 million Series D, nearly tripling its worth just six months after raising $75 million in Series C funding. The round, led by BOND with participation from CapitalG, Premji, and existing investors such as Greylock, Spark, IVP, Conviction, and 01a, brings the company’s total funding to over $285 million. The raise underscores the critical role inference now plays in bringing AI models into production at scale, positioning Baseten as one of the leading infrastructure providers for generative AI applications.
Founded in San Francisco by Tuhin Srivastava, Amir Haghighat, and Philip Howes, Baseten has focused from the start on building infrastructure specifically optimized for inference. The company has reached key milestones quickly: powering billions of weekly model calls for healthcare, enterprise, and consumer AI products; introducing Baseten Model APIs and Baseten Training to accelerate deployment and customization of models; and securing major customers including Abridge, Clay, Captions, OpenEvidence, and Writer. Its technology is already embedded in mission-critical use cases such as generating millions of clinical notes per week for healthcare providers.
With Series D funding, Baseten plans to expand applied research, infrastructure, and developer tooling, while strengthening its customer teams. CEO Tuhin Srivastava framed the company’s ambition: “If cloud was the foundation that enabled the latest generation of great technology companies, inference is the foundation for the next.”
• $150 million Series D funding round led by BOND; valuation rises to $2.15 billion
• Founded by Tuhin Srivastava, Amir Haghighat, and Philip Howes in San Francisco
• Powers billions of model inference calls weekly across healthcare and enterprise AI
• Customers include Abridge, Clay, Captions, OpenEvidence, and Writer
• Milestones include launch of Baseten Model APIs and Baseten Training
• Total capital raised now exceeds $285 million
🌐 Analysis: Baseten’s trajectory reflects how purpose-built inference platforms are moving from niche infrastructure to essential components of AI adoption. The founders’ backgrounds span applied machine learning, infrastructure engineering, and developer platform design—giving the company a multi-disciplinary edge. Milestones such as supporting large healthcare deployments demonstrate enterprise-grade reliability, while new developer APIs and training services broaden the stack. While Baseten has not disclosed hardware partners, its positioning suggests close alignment with GPU-accelerated environments common in hyperscale AI deployments. Against competitors like Together AI and Anyscale, Baseten’s focus on inference as the “app layer” of AI sets a clear differentiation as it builds out its ecosystem.
