Thinking Machines Lab inks massive compute deal with Nvidia

Thinking Machines Lab, an AI research startup co-founded by OpenAI alum Mira Murati, has signed a significant deal with semiconductor giant Nvidia to deploy at least one gigawatt of Nvidia’s new computing technology starting in 2027. Although the financial terms were not disclosed, Nvidia is also making a strategic investment in the company, which has raised over $2 billion since its founding in February 2025 from investors including Andreessen Horowitz, Accel, Nvidia, and AMD’s venture arm. The partnership aims to develop AI training and serving systems optimized for Nvidia’s architecture, with Murati emphasizing the foundational role of Nvidia’s technology in advancing AI development. Thinking Machines Lab, still in its early stages without any released products, focuses on building AI models that produce reproducible results. The company has experienced notable talent departures recently, including co-founder Barret Zoph and others returning to OpenAI, and another co-founder moving to Meta. This deal reflects the intense demand for large-scale compute resources in