Fractile Bags $220M to Compete With Nvidia in AI Hardware

British AI chip startup Fractile has secured $220 million in Series B funding, backed by Accel, Founders Fund, and Factorial Funds, to challenge Nvidia in AI inference.

British AI chip startup Fractile has closed a $220 million Series B funding round. The round was co-led by Accel, Factorial Funds, and Peter Thiel’s Founders Fund. The deal marks a major milestone for the young company. Additionally, existing backers Kindred Capital, the NATO Innovation Fund, and Oxford Science Enterprises who co-led Fractile’s $15 million seed round in July 2024 also participated in this latest round.

Furthermore, the round drew high-profile individual interest. In January 2025, former Intel CEO Pat Gelsinger said he had personally invested in the startup.

Fractile is designing AI compute systems aimed at making frontier model inference 25 times faster, at one-tenth the cost. In other words, the company is targeting the deployment side of AI not the training side, where Nvidia currently dominates.

Conventional AI accelerators, including Nvidia’s H- and B-series GPUs, separate the compute die from high-bandwidth memory. This separation creates an energy and latency penalty as data moves between them. Fractile is taking a different path. Its design instead performs the matrix multiplications that dominate transformer inference directly inside SRAM cells located alongside the compute logic.

The company calls this an in-memory-compute approach. It says the method removes most of the DRAM dependence that currently drives up inference costs. Fractile was founded in 2022 by Walter Goodwin, a 28-year-old PhD graduate from the University of Oxford’s Robotics Institute.

The team includes engineers drawn from Graphcore, Nvidia, and Imagination Technologies. The company is also building its own software stack alongside the silicon. Fractile currently has offices in London and Bristol in the UK, San Francisco in the U.S., and Taipei in Taiwan.

As AI models become more widely deployed, inference the process of running a model to generate outputs is now the dominant cost for AI companies. Consequently, purpose-built inference chips are a growing market.

Fractile is part of a small group of European chip startups arguing that the inference market is structurally distinct from training, and therefore winnable. Their argument is that training will continue to require the largest, most exotic systems and that Nvidia’s CUDA advantage is strongest there. Inference, by contrast, rewards specialized architectures tuned for throughput and energy per token, rather than peak processing power.

Interest from major AI labs is already building. A recent report from The Information claimed that Anthropic held discussions with Fractile regarding the purchase of its inference chips when the hardware becomes available in 2027.

If the relationship formalizes, Fractile would become Anthropic’s fourth named compute supplier, alongside Nvidia, Google’s TPUs, and Amazon’s Trainium and Inferentia chips. The funding is also being celebrated at a political level. UK ministers hailed the raise as a “vote of confidence” in Britain’s AI ambitions.

ALSO READ : Jensen Huang’s Foundation Donates $108M in AI Computing to Researchers

Moreover, the round follows Fractile’s February announcement of a £100 million, three-year expansion of its London and Bristol operations, including a new hardware engineering site in Bristol. The plans fit the broader UK sovereign-AI push that has also produced partnerships between BT, Nscale, and Nvidia on data centers.

Tape-out and customer integration are the next visible milestones for the company. Ultimately, Fractile is betting that the AI hardware race will not be won on training power alone but on who can serve intelligence cheapest and fastest at scale.

Disclaimer: This news is based on publicly available information. NervNow has not independently verified any claims.

MORE ON AI STARTUPS & FUNDINGS
Dessn Bags $6M to Help Designers Collaborate Directly in Codebases
Vapi Hits $500M Valuation Following $50M AI Funding Round
Helsing Could Become Europe’s Most Valuable Private Defense Tech Startup
Checkstep Raises £3M to Scale AI Content Moderation Globally