London-based AI inference chip startup OLIX raises $312 million Series B at $3.3 billion valuation to scale production and deployment of its inference accelerator platform
London-based AI chip startup OLIX, founded by James Dacombe, has raised $312 million in a Series B funding round, valuing the two-year-old company at $3.3 billion. The round was led by Fundomo, Arm, and Hudson River Trading, with Netflix co-founder Reed Hastings participating as an angel investor. Existing investors—Hummingbird Ventures, Crane, Plural, Creandum, Phoenix Court, and Transition—also increased their commitments.
OLIX is developing an AI inference platform that departs from the industry's reliance on general-purpose processors. Rather than designing a single chip to handle every aspect of an AI workload, the company is building specialized processors for different stages of token generation. The company argues that running every stage on the same general-purpose processor limits efficiency, and instead advocates dividing the workload across multiple specialized chips, each concentrating on a particular part of the model.
This approach forms the basis of the X-1 platform. Models are distributed, or "unrolled," across a large number of chips in a production-line arrangement. Each chip handles a defined section of the model while retaining a flexible computing architecture, rather than being permanently designed around one specific AI model. OLIX is designing the chips, lasers, and networking components used in the system, with the finished infrastructure supplied as complete racks for demanding AI inference workloads.
The X-1 platform uses what OLIX describes as a "slow and wide" optical interconnect, replacing copper connections with direct light-based data transfer between chips. The company claims this architecture reduces latency and energy consumption. The platform is being developed through rack-scale codesign, bringing chips, optical links, and broader system components together rather than treating them separately. Workloads will be allocated across racks through a deterministic compiler.
The first processor planned for the X-1 platform is DX-1, a decode accelerator designed for the stage where an AI model reasons and produces its response. For models with 100 billion parameters, OLIX claims DX-1 can deliver more than 10,000 tokens per second per user while achieving higher output-token throughput per watt than general-purpose chips operating with large batch sizes. The multi-rack architecture is designed to support models with 10 trillion parameters or more. DX-1 stores models in fast on-chip SRAM to improve energy efficiency and reduce latency, and has been designed without advanced packaging or high-bandwidth memory—components in limited supply across the semiconductor industry—to make it easier to increase production volumes.
OLIX plans to use the Series B proceeds to deliver DX-1 to first customers during the second half of 2027, while also supporting development of the wider custom silicon platform and manufacturing arrangements needed to produce hardware at scale. The company is recruiting across silicon, photonics, compiler, and systems engineering in London, Bristol, Austin, Toronto, and San Francisco.