River AI, a startup focused on personally trainable AI agents, raised $1.1 billion in a seed/Series A round announced in August 2026 — just two months after the company emerged from stealth in June 2026.
The round was led by General Catalyst and AMP PBC, with participation from Nvidia, AMD Ventures, Y Combinator, and Temasek. AMP PBC is an AI-focused investment firm founded in 2026 by Anjney Midha, a former general partner at Andreessen Horowitz who backed companies including Black Forest Labs, Mistral AI, LMArena, and OpenRouter.
River was founded by Igor Babuschkin, a co-founder of xAI whose previous roles include AI positions at DeepMind and OpenAI. Babuschkin’s stated goal is to rebuild the AI stack from the ground up — covering training, models, the product layer, and hardware — in order to create agents that are personally trainable rather than designed as general-purpose worker replacements. “Capable agents will be a normal part of everyday life,” he wrote at launch. “Less like the assistants you call on today when you need a task done, more like guardian angels: quietly present, on your side, helping with what actually matters to you.”
River’s first product is an API, billed per one million tokens, that allows developers to apply reinforcement learning and low-rank adaptation fine-tuning to open models. The company positions this as an alternative to prompt engineering, letting users train and serve models as their own. For enterprises, River claims complex reinforcement learning runs can be completed in 15 to 20 minutes with no infrastructure team, at two to four times the cost savings compared to closed-source alternatives.
The size of the raise may signal growing enterprise appetite for controlling AI model deployment through open-weight models. River’s neocloud offering targets the post-training expertise gap that enterprises face as they move toward mixed-model strategies.
Source: TechCrunch