Tendrils Compute Raises $30M to Rethink the CPU
Cambridge startup Tendrils Compute has raised $30m, led by Benchmark, to build a parallel-first processor based on interaction nets rather than conventional CPU design.

Tendrils Compute, a Cambridge-based chip startup, has raised $30m in a funding round that was reportedly hotly contested. Benchmark led the round, which aims to back a radically different kind of processor.
The company wants to challenge the dominance of conventional CPU and GPU architecture. Tendrils believes general-purpose chips built on its computing model, called interaction nets, can scale without the cost difficulties and long-standing design problems tied to today's CPUs. Its technology is meant to run thousands of small processes at the same time and handle complex workloads.
A parallel-first design
The company describes its approach as a break with sequential design. "Instead of starting from a sequential design, we're building our processor to have thousands of cores operating in parallel," Tendrils said. That requires moving beyond shared mutable memory and linear instruction sequences, it added.
Alongside the hardware, Tendrils is working on Vine, an open-source programming language designed from the start to target a new wave of parallel-first architectures.
Investors and open questions
Besides Benchmark, the round includes Visionaries Clubs, the UK Sovereign AI Fund, FR8, Antimatter, Cloudberry VC, Conviction, Embassy Ventures and SV Angels. Tendrils is one of several UK chip companies, along with Olix and Fractile, that investors are now backing.
The key caveat is performance: the startup has not yet disclosed any hardware benchmarks, so there is no public evidence of how its chips compare with existing processors. For now, the funding signals investor appetite for alternatives to established architectures rather than proof that the approach works at scale.
Apps worth a look
Tap the right tiles, add them up, hit the target. Easy to learn, hard to master. Free, for all ages.
Talk to an AI conversation partner; every sentence comes back corrected and explained in your language. 9 languages, A1 to C2.
Open EDIFACT and X12 messages, see every segment explained, convert to JSON/XML, export PDF and Excel. Runs locally, no upload.
Pick who it is for, the occasion and their personality. The search engine reads product descriptions from local shops and ranks what fits. Free, no sign-up.
More from this day
See all →More in: Hardware
See all →
QuantumDiamonds: Munich Start-up Uses Diamond Sensors to Find Chip Defects

AMD Epyc Venice-X Brings Back 3D V-Cache With 5.3 GHz Clocks

Seagate FireCuda Forge Brings Smart Caching, Up to 20TB




