MatX Secures $500M Series B to Face NVIDIA Head On in AI Training Chips

AI semiconductor startup MatX has secured $500 million in Series B funding to accelerate development of processors designed to outperform Nvidia GPUs in large language model training. The round was led by Jane Street and Situational Awareness, an investment fund founded by former OpenAI researcher Leopold Aschenbrenner, with participation from Marvell Technology, NFDG, Spark Capital, and Stripe co-founders Patrick Collison and John Collison. Founder and CEO Reiner Pope, previously head of AI software for Google’s TPU program, co-founded the company in 2023 with former TPU hardware lead Mike Gunter.

MatX aims to deliver chips capable of training AI models significantly faster than existing hardware. The new capital will support manufacturing through TSMC, with initial shipments targeted for 2027, positioning the company as a major emerging competitor in the AI compute market.