Fractile Raises $720M Series C to Take On the AI Chip Giants
Fractile's $720M Series C, backed by Britain's Sovereign AI Fund, values the AI chip startup above £5 billion (~$7B). First chips and systems ship next year.
By Nathan Brooks
2 min read

What's News
- Fractile raised a $720 million Series C at a valuation above £5 billion (~$7 billion).
- The round was led by Thrive Capital, Founders Fund, Lightspeed Venture Partners and Redpoint.
- Britain's Sovereign AI Fund backed the round as part of a U.K. hardware strategy.
- CEO Walter Goodwin says the funding will let Fractile ship its first chips and systems next year.
- The design stacks DRAM on top of each chip to bypass the memory-capacity tradeoff.
British AI chip startup Fractile has closed a $720 million Series C at a valuation of "over £5 billion" — roughly $7 billion, per figures reported by the Financial Times and a Reuters source. Britain's Sovereign AI Fund backed the round, and the government in London framed the deal as a way of "ensuring the U.K. has a greater stake in the hardware that will underpin its future economy."
Thrive Capital led the financing, alongside Founders Fund, Lightspeed Venture Partners and Redpoint. The scale of the check puts Fractile in the top tier of funded challengers building purpose-built AI silicon, in the company of Axelera, Etched and OLIX.
Why inference chips are suddenly a $700 million bet
The whole category exists to attack one cost: inference. That's the price of running increasingly large AI models in production, and it's climbing as models get hungrier. Purpose-built chips are one of several routes companies are exploring to bring that bill down.
Demand has intensified with the rise of agentic AI. Companies now want models to run long chains of tasks without losing speed, which strains conventional compute. Hopes are high for dedicated AI chips, even though supply has been limited so far.
What does Fractile's design actually change?
For practitioners, the technical story matters more than the fundraising headline. Fractile's differentiation lives in its hardware design, which the company arrived at after a pivot.
The approach revolves around DRAM — the most common type of RAM — but stacked on top of each chip in a configuration the company believes will bypass the usual tradeoff between memory and compute capacity. That tradeoff has become a genuine bottleneck for agentic workloads, where memory bandwidth dictates how fast a large model can keep working through extended task chains.
Fractile isn't stopping at silicon, though. It's taking a full-stack approach that includes a machine learning compiler, built to maximize the speed at which LLMs execute on its chips. That's why the company frames itself not as a chipmaker but as a frontier AI company — and why investors priced the round at a frontier-grade valuation.
What the founders are saying
CEO Walter Goodwin struck a measured tone in a LinkedIn post following the announcement. "There is still a lot of work ahead," he wrote, while confirming that the new funding will let the startup "ship [its] first chips and systems next year."
That timeline is the number practitioners should hold onto. A valuation north of £5 billion — call it roughly $7 billion — rests on hardware that hasn't shipped yet, in a market where incumbents and rivals alike are racing to solve the same inference-cost problem.
For the U.K., the deal marks its most significant positioning yet in the AI hardware supply chain. For Fractile's customers-to-be, next year's first silicon will be the moment the DRAM-stacking thesis gets tested against real workloads.
Original: linkedin.com
More from Nathan Brooks
Show full bio
News editor covering regulation, policy and their impact on entrepreneurship, startups and venture funding at Business Bearings.
2,030 articles