Thinking Machines Lab, the AI startup founded by former OpenAI executive Mira Murati, confirmed a multi-year strategic partnership with Nvidia on Tuesday to secure a massive supply of next-generation semiconductor hardware.
A Strategic Bet on Vera Rubin Systems
As part of the agreement, Thinking Machines Lab will deploy at least one gigawatt of Nvidia’s Vera Rubin systems. The rollout of this advanced architecture is scheduled to begin in 2027. While the financial specifics of the deal remain undisclosed, the partnership marks a significant escalation in the race for AI infrastructure.
Nvidia is also taking an equity stake in the startup. Thinking Machines Lab has already secured over $2 billion in funding since its February 2025 inception, backed by heavyweights such as Andreessen Horowitz, Accel, and even the venture arm of rival chipmaker AMD.
Scaling Reproducible AI
Currently valued at over $12 billion, the seed-stage company is focused on developing AI models designed for reproducible outcomes. In October, the firm launched its first commercial product, an API titled Tinker.
Beyond hardware procurement, the multi-year strategic partnership with AI semiconductor giant Nvidia includes a commitment to co-develop training and serving systems specifically optimized for Nvidia’s architecture.
“Nvidia’s technology is the foundation on which the entire field is built,” Mira Murati stated in the official announcement. “This partnership accelerates our capacity to build AI that people can shape and make their own, as it shapes human potential in turn.”
Leadership Shifts and Industry Demand
The deal arrives during a period of transition for the young startup. Thinking Machines Lab has experienced several high-profile departures recently; co-founder Andrew Tulloch left for Meta in October, while three other co-founders—Barret Zoph, Luke Metz, and Sam Schoenholz—returned to OpenAI earlier this year.
The aggressive push for compute power aligns with broader industry trends. Nvidia CEO Jensen Huang has projected that total corporate spending on AI infrastructure could reach $3 trillion to $4 trillion by the end of the decade. This deal follows other historic industry movements, such as OpenAI’s reported $300 billion compute agreement with Oracle in 2025.
Representatives for both Thinking Machines Lab and Nvidia declined to provide further details regarding the terms of the investment or the hardware agreement.
