AI chip startup MatX has raised $500 million in a Series B funding round led by an investment fund founded by a former OpenAI researcher.

Leopold Aschenbrenner’s Situational Awareness co-led the round with Jane Street. Other investors include Spark Capital, Triatomic Capital, Harpoon, Alchip Technologies, and Marvell.

Launched in 2024 by former Google engineers Reiner Pope and Mike Gunter, MatX is aiming to build processors specifically designed to support large language models (LLMs). While working at Google, Pope wrote AI software while Gunter designed hardware, including chips, for the software to run on.

In a LinkedIn post announcing the fundraise, Pope said the company’s forthcoming MatX One chip will deliver “much higher throughput than any other chip while also achieving the lowest latency.”

Based on a splittable systolic array, a chip architecture that breaks processing elements into smaller systolic arrays to help boost efficiency, Pope said the chip will combine the low latency of SRAM-first designs with the long-context support of HBM.

“These elements, plus a fresh take on numerics, deliver higher throughput on LLMs than any announced system, while simultaneously matching the latency of SRAM-first designs. Higher throughput and lower latency give you smarter and faster models for your subscription dollar,“ Pope wrote.

According to TechCrunch, MatX will produce the chips with TSMC and plans to start shipping them in 2027.