Igor Babuschkin's River AI raises $1.1B to build an open AI stack
What's the deal? River AIDealroom has a profile for this one. Try Dealroom →, the full-stack AI company founded by xAIDealroom has a profile for this one. Try Dealroom → co-founder Igor BabuschkinDealroom has a profile for this one. Try Dealroom →, has raised $1.1 billion in early-stage funding. General Catalyst and AMPDealroom has a profile for this one. Try Dealroom → PBCDealroom has a profile for this one. Try Dealroom → led the round, with strategic investment from NVIDIA and AMD VenturesDealroom has a profile for this one. Try Dealroom →, plus Y Combinator and Temasek.
Why it stands out: The round ranks in the top 1% of all early-stage VC deals ever recorded in its industry and region, based on a sample of 3,409 rounds — an unusually large sum for a company at this stage.
What's the endgame? River is building a full stack: hardware that keeps personal AI close to users, training infrastructure that makes fine-tuning accessible, and products built around personalisation. Its API delivers LoRA fine-tuning and reinforcement learning for open-weight models, handling infrastructure so developers can focus on their models.
The company says any enterprise can complete a reinforcement learning training run in 15 to 20 minutes without an infrastructure team, at two to four times the cost savings versus closed-source alternatives. Billing is metered on tokens used, eliminating the cost of idle GPU capacity.
The New York Times reported that Babuschkin wants River to build servers that let individuals and businesses run open-source AI on their own hardware, keeping private data away from River AI or any other company. He also envisions a digital assistant that works across devices and can be retrained or modified by users. Based in Palo Alto, River AI has about 20 employees and is still months away from releasing its open-source technologies.
What they said: "AI should be open, freely available, and affordable. It should feel like it is working for the person using it, not the lab that trained it," said Babuschkin, co-founder and chief executive officer. "We started River to allow people and companies to own their intelligence."
Who's behind it: Babuschkin previously worked on generative modelling and reinforcement learning at Google DeepMind, led large-scale training at OpenAI, then co-founded xAI. River's founding team brings experience from xAI and Tesla.
The signal: River's backers frame open models as a national priority. "American leadership in AI urgently requires leadership in open weight models, while maintaining a lead in closed frontier models," said Hemant Taneja, chief executive of General Catalyst. The bet is that ownership of intelligence shifts from the labs that train models to the people and companies that use them.
Image credit: River AI
Read more: Associated Press · The New York Times
