The WSJ writes that AI companies can't buy or build datacenter capacity fast enough to keep up with explosive demand growth. Some of the LLM/VLM model processing MUST be pushed onto devices. Programmable edge devices that can run fast-changing 30B -> 70B models are needed. See how Quadric #Chimera AI processors address the LLM/VLM challenge. https://lnkd.in/gg-vbZhY https://lnkd.in/gXafeJWz
About us
Quadric licenses an AI processor architecture optimized for on-device inference. Only the Quadric Chimera GPNPU (general purpose neural processing unit) delivers high AI/ML inference performance while also running C++ code without forcing the developer to artificially partition application code between two or three different processors. Quadric's Chimera GPNPU processor IP core scales from 1 to 864 TOPS and seamlessly intermixes scalar, vector and matrix code.
- Website
-
http://quadric.ai
External link for Quadric
- Industry
- Semiconductor Manufacturing
- Company size
- 51-200 employees
- Headquarters
- Burlingame, CA
- Type
- Privately Held
- Founded
- 2017
Locations
-
Primary
Get directions
Burlingame, CA 94010, US
Employees at Quadric
Updates
-
If you missed Nigel Drego's presentation at Ride AI in San Francisco last week, you can watch it on YouTube: https://lnkd.in/gRV2Z69z
The Last Mile Isn't Software, It's Silicon: Quadric Demo -- Ride AI 2026
https://www.youtube.com/
-
Demand for Tokens - by the AI Agents that gobble them up 24/7 - is spiking demand for edge compute devices that can run LLMs, including a run on $2000 Mac Minis (per WSJ). The rush is on to find ways to deliver programmable token engines on the edge at a far more economical price point. Find out how #Chimera processors can help. https://lnkd.in/gg-vbZhY https://lnkd.in/gPr8XHr4
-
Join Quadric at Microeletronics US and discover Quadric Chimera™ programmable edge AI, powering next-gen devices. quadric.ai/events #QuadricAI #EdgeAI #OnDeviceAI #AIEdgeComputing #EnergyEfficientAI #HighPerformanceAI #MicroelectronicsUS #NextGenAI
-
-
At Quadric we don't just deliver world-class AI Inference processing solutions; we are also building and using cutting-edge AI tooling inside the company to develop and deliver our #Chimera AI processor IP. https://lnkd.in/gGr2x7c4
-
The AI revolution is going on-device! Meet Quadric at IP SoC and discover how LLMs run fast + efficiently at the edge quadric.ai/events #OnDeviceAI, #EdgeAI, #AIProcessors, #LLM, #AIatTheEdge, #NextGenAI, #NPU IP
-
-
If you're working in AI, semiconductors, or edge computing, this will be a great evening of ideas and discussion. Discover Quadric Solutions www.quadric. #AI #Semiconductors #EdgeAI #ComputerArchitecture #MachineLearning #HardwareDesign
-
-
How do you know an "AI processor" is a real processor, and not just a matrix multiplier with a fancy marketing name? If your #NPU cannot do this, your SoC project might be #DOOMed https://lnkd.in/gUQU56jm
-
Will you be visiting Embedded Vision Summit? Have an Edge AI project in development? Schedule a meeting with a Quadric SME to discuss your architecture, performance goals, and deployment strategy. https://quadric.ai/contact #Events #Solutions #EVS26
-
-
How Quadric is Enabling Dramatic Improvements in Edge AI Quadric CEO Veer Kheterpal explains how programmable NPUs, unified AI architectures, and better software are enabling dramatic improvements in edge AI performance, efficiency, and flexibility. If you’re building AI at the edge, this conversation is worth your time. Listen to the Podcast: https://lnkd.in/gz-TPjJ6
-