Mirendil raises $1 billion Series A to make frontier AI research cheaper and faster

Mirendil has raised $1 billion in a Series A round led by Kleiner Perkins and Clearlake Capital, with Google, Battery Ventures, Doerr Capital, M13 and Pear VC among the other investors, the company said in a press release. It did not give a valuation.

The company is led by CEO Behnam Neyshabur and CTO Harsh Mehta, who helped launch Anthropic's effort to automate parts of its internal research, and raised a $200 million seed round at a $1 billion valuation in June. Its aim is to let small teams do frontier AI research by handing the hands-on work to AI agents.

Mirendil says its first results show what that looks like. One researcher working with agents pretrained Mirendil-Base, a competitive coding model, from scratch in 3.5 months with less than $10 million of compute. The company then post-trained it into Fern, which can hold up to 10 million tokens of context, enough for days of an agent's work. Its product Lor, now in private beta, runs hundreds of agents that write code and run experiments for days at a time, and Mirendil uses it internally every day.