River AI, a two‑month‑old artificial‑intelligence startup, closed a $1.1 billion seed round on Tuesday. General Catalyst and AMP PBC led the financing, while Nvidia, AMD Ventures, Y Combinator and Temasek participated. The capital infusion positions River among the most heavily funded early‑stage AI firms.
Founder Igor Babuschkin—who co‑founded xAI and previously held senior roles at DeepMind and OpenAI—unveiled the company’s mission in a launch blog posted in June, when River emerged from stealth. Babuschkin argues that the prevailing AI trajectory, which focuses on replacing human workers, overlooks a more personal use case: agents that learn directly from their owners.
“To get there, we believe the stack has to be rebuilt end‑to‑end: training, models, the product layer, and new hardware that lets personal AI live close to you,” he wrote. The vision is of “guardian‑angel” assistants that know their users intimately, rather than the generic task‑oriented bots that dominate today’s market.
River’s first offering is an API that charges per million tokens, with rates varying by the open model chosen. Developers can apply reinforcement learning (RL) and low‑rank adaptation (LoRA) fine‑tuning to open‑source models, a move the company says will eliminate the need for cumbersome prompt engineering. “Prompting steers a model you don’t own and can’t improve. River lets you train open models into ones that are truly yours—and serve them like any other endpoint,” the product literature claims.
The round’s size is striking for a company barely a quarter of a year old, but it reflects a broader market shift. Enterprises are increasingly eager to control their AI destiny, mixing proprietary and open‑weight models. River promises to simplify the post‑training phase with a “neocloud” service that lets any organization run complex RL jobs in 15‑20 minutes without an internal infrastructure team, delivering two‑to‑four‑fold cost savings versus closed‑source alternatives.
AMP PBC, an AI‑focused investment firm founded in 2026 by former Andreessen Horowitz partner Anjney Midha, joined the lead investors. Midha’s track record includes backing firms such as Black Forest Labs, Mistral AI, LMArena and OpenRouter, underscoring the strategic heft behind the deal.
River’s broader ecosystem already shows signs of alignment. Nvidia, a participant in the round, is partnering with PC makers like Dell, Microsoft and HP to ship AI‑capable hardware. While River’s technical differentiation remains to be proven in the field, the company now has a war chest large enough to experiment aggressively.
Analysts note that the funding underscores the overheated atmosphere surrounding generative AI, yet River’s focus on personal, trainable agents could carve out a distinct niche. If the startup can deliver on its promise of a fully rebuilt stack, the $1.1 billion could translate into a new class of AI products that sit on users’ desks—or even on their laptops—rather than in distant data centers.
This article was written with the assistance of AI.
News Factory APP - agentic news to boost your SEO & AEO.