General Catalyst leads $1.1B round into 2-month-old River AI
| Source: TechCrunch AI
Tags: River AI, Igor Babuschkin, General Catalyst, open-source, fine-tuning, AI agents, funding
River AI, a 2-month-old startup founded by xAI co-founder Igor Babuschkin, raised $1.1B in seed/Series A from General Catalyst, Nvidia, AMD Ventures, and others to let individuals and enterprises train their own open-weight AI models—no infrastructure team required.
Details
River AI emerged from stealth in June 2026 and has secured $1.1 billion in a seed/Series A led by General Catalyst and AMP PBC (a new firm from former Andreessen Horowitz GP Anjney Midha), with Nvidia, AMD Ventures, Y Combinator, and Temasek participating. The company was founded by Igor Babuschkin, whose prior roles include AI research at DeepMind and OpenAI, and co-founding xAI with Elon Musk.\n\nRiver's core thesis: current AI is wrong because models are owned by companies rather than users. Their product is an API that lets anyone run reinforcement learning or LoRA fine-tuning on open models—billed per million tokens, with RL runs completing in 15–20 minutes at 2–4x cost savings versus closed-source alternatives. No infrastructure team required. The longer-term vision is 'guardian angel' personal agents that 'know you well, and will be yours, not someone else's.'\n\n$1.1B for a two-month-old company is extraordinary even by 2026 AI standards. Hardware co-investors Nvidia and AMD Ventures suggest River is building custom compute pathways for personal agent inference. This positions River squarely against OpenAI's fine-tuning business and the closed-model ecosystem.