General Purpose Neural Processing Unit

A high-performance Machine Learning processor and a C++ programmable DSP in one unified architecture
the quadric Architecture

One Architecture.

Every Algorithm.
Modern System-on-Chip architectures deploy complex algorithms that mix traditional C++ based code with newly emerging and fast-changing machine learning (ML) inference code.  Only Quadric’s GPNPU architecture can deliver high ML inference performance and run complex C++ code on the same fully programmable processor without forcing the developer to artificially partition code between two or three different kinds of processors.  Quadric’s GPNPU is a licensable processor that scales from 1 TOP to 64 TOPs and seamlessly intermixes scalar, vector and matrix code.
Explore the Architecture
Die photograph of the first quadric Processor: the q16.
Silicon Proven
Not only is our architecture built from scratch to meet the needs of complex workloads, it is silicon-proven. Our proof of concept demonstration chip runs at 1 GHz and serves as a validation vehicle for the architecture and our complete software stacks.
Learn about our Processor
Limited Availability

Early Access Partner

Development Kit
Quadric’s first generation GPNPU development kit works with any M-key M.2 Host at speeds up to PCIe Gen4.
Learn more about the Dev Kit
Close up photograph of the quadric Dev Kit v1.0. The easiest way to experience the q16 Processor.
Register Your Interest
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Quadric Developer Studio
The quadric SDK gives developers a powerful tool for the effortless simulation and deployment of AI software on any product containing a quadric architecture instance. Write your algorithms once and have the peace of mind to deploy them everywhere.

We've released a hosted version of the SDK on our Developer Studio.
Experience the Developer Studio

Hosted SDK

Developer Studio Limited Preview
The latest insights from our team
In the News

We've built our software using the power of open source.

We stand on the shoulders of giants.

We can reach higher by leveraging the ecosystem.

tvm logotensorflow logopytorch logoLLVM logoDocker logo