P2Matrix

Compute-in-memory silicon for AI inference.

P2Matrix puts the arithmetic beside the memory that holds the model, so the weights never leave the chip. Our investor data room is available by request.

Request access

Tell us who you are. We review every request and email an access code to approved investors.

Have an access code?

Sign in with the email address you requested access with.

© 2026 P2Matrix · contact@p2matrix.ai