Thinking Machines in talks for $1B at $40B valuation as Mira Murati’s AI lab races ahead
Tech Funding News Abhinaya Prabhu ● Covered by 2 sources
Thinking Machines is trying to raise $1B at a $40B value. That’s way below the $50B talk that blew up after key exits.
Based on reporting by Tech Funding News, Abhinaya Prabhu — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
Thinking Machines is back in the market, this time trying to raise $1 billion at a valuation of at least $40 billion. Accel is in talks to lead the round, and Nvidia may join as well. It’s a sharp step down from the $50 billion to $60 billion talks that surfaced in November and then collapsed in January.
That reset matters because the company has been through a very public wobble. In July 2025, it closed the largest seed round in AI history: $2 billion at a $12 billion valuation, with Andreessen Horowitz leading and Nvidia, GV, Lightspeed and Conviction Partners joining. Back then, investors seemed to be buying the aura around Mira Murati, the former OpenAI CTO, plus the roughly two dozen OpenAI veterans who came with her.
The product story has moved since then, but only so far. In July, the company shipped Inkling, an open-weight model tied to Tinker, where customers pay usage-based compute fees to adapt the model to their own data. Annualised revenue now tops $100 million, according to a source familiar with the finances. On paper, that puts the company above 400 times revenue.
That is the kind of number that can make even AI investors blink, though apparently not for long. Accel has already committed roughly $8.5 billion across new funds this year, and Thinking Machines remains one of its marquee bets. Nvidia’s possible participation also fits the company’s setup: Thinking Machines signed a multiyear partnership with the chipmaker in March, with at least one gigawatt of upcoming Vera Rubin systems and an undisclosed investment attached.
The bigger question is whether the market wants to pay for the same story twice. The company’s valuation talks fell apart after leadership turmoil, including the January exit of co-founder and CTO Barret Zoph and later the departure of founding researcher Lilian Weng. Now it is asking investors to believe that a product, some revenue and a lot of compute can carry a much richer price tag than the last time around.
My take — AI-written commentary, not fact-checked reporting
This is classic AI-era pricing: a revenue number gets waved around like a shield, and everyone pretends 400x is a normal way to run a market. It isn’t. If a company’s main edge is access to compute and famous alumni, the bill should be paid in caution, not celebration.
Read more about this at: Tech Funding News