The news
Etched, a startup that builds hardware for running AI models, is reviewing investment offers that would value it at $40 billion or more, TechCrunch reported on October 5, 2026, citing people familiar with the company. The bids would roughly double the valuation Etched set in its most recent round.
According to TechCrunch, the incoming bids range from $40 billion from top-tier investors to $50 billion from lesser-known backers. The outlet described the discussions as early talks. Etched declined to comment, and no new round has been announced.
The offers follow two large raises in quick succession. In July 2026, Etched raised $300 million at a $10.3 billion valuation in a round led by Sequoia Capital, according to TechCrunch. It then raised $700 million at a $21 billion valuation in a round led by Jane Street, the quantitative trading firm that is also a customer. VKTR dated that announcement to August 18, while TechCrunch's report places the round in September.
Etched focuses on inference, the work of running already-trained AI models to answer requests, rather than training them. TechCrunch reported that the company builds full hardware systems around its own chips, designed two new components from scratch to speed up inference, and made a test chip at Taiwan Semiconductor Manufacturing Co. (TSM) this summer. Etched claims its processors can handle more tokens faster and at lower cost than Nvidia (NVDA) chips.
The company has about 400 employees, roughly 15% of whom previously worked at Nvidia, according to a Wall Street Journal figure cited by TechCrunch. It operates a 10-megawatt data center in Silicon Valley. VKTR reported that Etched had raised $1.9 billion in total and had more than $1 billion in booked customer contracts at the time of the Jane Street round, and that Jane Street received its first rack in July.
The numbers
- Reported bid range
- $40 billion to $50 billion
- Last valuation (Jane Street-led round)
- $21 billion
- Jane Street-led round
- $700 million
- July 2026 round (Sequoia-led)
- $300 million at $10.3 billion
- Employees
- About 400
- Total raised (per VKTR)
- $1.9 billion
Why CEOs should care
For technology buyers, the size of these bids is a signal that well-funded alternatives to Nvidia for inference are likely to keep coming. Inference is where most companies' AI compute bills grow once models move into production. Teams negotiating multiyear GPU or cloud commitments should ask whether contracts leave room to shift workloads to specialized hardware if the price-performance claims hold up.
For CFOs and investors, the speed of the markups is the point to watch. A company moving from $10.3 billion in July to bids of $40 billion or more within months is being priced on expected future demand, not current scale. Those claims about tokens, speed and cost come from Etched itself; independent benchmarks and customer deployments beyond Jane Street are what would confirm them.
For boards weighing AI infrastructure strategy, the main takeaway is concentration risk. Heavy capital flowing into challengers suggests investors expect inference spending to be large enough to support more than one major supplier. Ask management how dependent the company's AI roadmap is on a single chip vendor and what it would take to test an alternative.
The bigger picture
Nvidia still dominates AI chips, but inference is drawing a wave of specialized competitors because it is a recurring cost that grows with every user request. Etched's bet is that purpose-built systems can beat general-purpose graphics processors on cost per answer. The capital chasing that bet has grown fast in 2026, with Etched's valuation roughly doubling between each round.
What’s next
Watch for whether Etched accepts a bid and at what price, whether it names new investors, and whether customers beyond Jane Street confirm production deployments of its systems.
What “Fact-checked” means
Fact-checking means testing a story’s facts against the evidence before it is published. This story went through at least two separate checks before this version was published.
- What we checked
- Its names, figures, dates, job titles, quotes and who said what were checked against the story’s sources, including its main source where it could be opened. The headline was checked for accuracy and overstatement.
- How
- A first check reviewed the whole story. If it passed, a second, skeptical check went back to the sources to look for mistakes in the most important facts. If a check flagged the story, it was edited to fix the problems found, and a separate re-check then reviewed the whole story again.
- Who
- The checks are made by our newsroom, as steps kept separate from the writing, under rules set by our editor, Hussein Mukhtar. A story the checks still flag is held for the editor, who decides whether it is fixed, published or dropped.
- If something is wrong
- “Fact-checked” does not mean error-free. If a material error is found after publication, we correct the story and add a note saying what changed. Report an error
Companies in this story
