Positron Develops Custom ASIC
Positron AI secures $51.6M Series A to advance cost-efficient AI chips, enabling 16T-parameter models & cutting GPU dependency. Atlas FPGA systems now available.
"ChipPub" Publication: 20% Discount Offer Link.
Positron recently announced the completion of an oversubscribed $51.6 million Series A funding round, bringing its total funding this year to over $75 million. The new capital will support the continued deployment of Positron AI’s first-generation product, Atlas, and accelerate the launch of its second-generation product in 2026.
Positron AI’s Vision for Cost-Effective AI
Mitesh Agrawal, CEO of Positron AI, stated: “We founded Positron to meet the demands of modern artificial intelligence: running cutting-edge models with the lowest per-token generation cost and the highest memory capacity. Our highly optimized chip and memory architecture enables superintelligence in a single system. Our goal is to run models with up to 16 trillion parameters, handling tens of millions of tokens in context length or memory-intensive video generation models.”