Mira Murati's Thinking Machines Lab raising $5 billion at $40 billion valuation
Mira Murati's Thinking Machines Lab is reportedly closing a $5 billion round at a $40 billion pre-money valuation.
Objective Facts
Business Insider reported on November 6, 2026 that Thinking Machines Lab, the AI startup founded by former OpenAI chief technology officer Mira Murati, is raising $5 billion at a $40 billion pre-money valuation in a round led by Accel that could close as soon as that week. AI Weekly, citing the same report, said Nvidia is separately in talks to put in about $2.5 billion. Thinking Machines declined to comment and Accel did not respond to requests, according to Business Insider. The new mark would more than triple the $12 billion valuation from the company's $2 billion seed round in July 2025, though it sits below the roughly $50 billion target from talks that Business Insider said were abandoned late last year. The funding report coincided with the company's release of Inkling 2 Small, a second generation open-weight model.
Deep Dive
The round caps a year of shifting numbers. TechCrunch's Marina Temkin reported on September 3 that Thinking Machines was in talks to raise $1 billion at a valuation of at least $40 billion, with Accel in talks to lead. Days later, The Next Web's Cristian Dina wrote that The Information had revised the figure to $5 billion to $6 billion, with Nvidia supplying roughly half. Business Insider's November 6 report settled on $5 billion at $40 billion pre-money, which Techstrong.ai and Tech Funding News put at about $45 billion after the money comes in. Nvidia already has a commercial tie to the lab: in March the two companies announced a deal to deploy at least one gigawatt of Nvidia Vera Rubin systems, with first deployment targeted for early 2027. The bull case rests on product momentum. SiliconANGLE's Maria Deutscher reported that Inkling 2 Small has 594 billion total parameters with about 17 billion active per prompt, and that it outscored several open-source rivals on knowledge-work benchmarks. GeekPark, in a piece republished by TechFlow, said Artificial Analysis rates it the highest-scoring open-weight model outside China, while still about 15 points behind Moonshot AI's Kimi K3.1 Pro. The skeptical case rests on revenue and staffing. TechCrunch put annual revenue above $100 million, and The Next Web estimated the valuation at roughly 80 to 400 times revenue, figures that are self-reported and unaudited. Techstrong.ai noted that the company has disclosed little about its finances, and Business Insider reported that several founding members, including Barret Zoph, Luke Metz and Andrew Tulloch, have left for Meta and OpenAI. Techstrong.ai also reported that Inkling 2 Small cost about six times as much per Artificial Analysis task as its predecessor. The next items to watch are whether the round actually closes on the reported terms, whether Nvidia's participation is confirmed and at what size, and whether Thinking Machines releases the Inkling 2 Small weights on schedule after external safety testing. Everything-PR's Sarel Doglu argued in October that reported valuations quickly become the reference point for later coverage whether or not a company confirms them, so an official announcement would be the first figure from the company itself.