Quadric
Company Information
Headquarters
Burlingame, California, USA
Year Founded
2017
Industries
Semiconductors, Automotive, Robotics, Consumer Electronics, Industrial Automation, Computer Hardware, Networking
What Is Quadric?
Quadric is an AI semiconductor technology company developing programmable processor IP for on-device AI inference. Its core technology, the Chimera architecture, is a General-Purpose Neural Processing Unit (GPNPU) designed for integration into system-on-chip (SoC) products. Quadric combines AI inference acceleration with general-purpose programmability, allowing chip designers to support neural networks, computer vision, signal processing, and other data-parallel workloads within a unified processor architecture.
What Does Quadric Do?
Quadric licenses AI processor intellectual property (IP) and provides the software tools needed to integrate and program that technology inside custom chips and SoCs. Rather than manufacturing finished AI chips for end users, Quadric gives semiconductor and system companies processor designs that can become part of their own silicon.
Its Chimera processors are designed to execute complete AI inference pipelines while also handling C++ algorithms that might otherwise require separate CPUs, DSPs, and neural-network accelerators. The technology can scale across different performance and power requirements, supporting on-device AI applications ranging from automotive systems to robotics, cameras, wearables, and industrial equipment.
Quadric Technology
Quadric’s core technology is the Chimera General-Purpose Neural Processing Unit (GPNPU). Unlike fixed-function NPUs that implement a predetermined set of neural-network operations in hardware, Chimera is programmable and can execute custom operators through C++. This allows developers to accommodate new AI models and operators through software rather than necessarily redesigning the underlying silicon.
The architecture combines data-parallel processing with an instruction-driven programming model. Quadric describes it as a hybrid data-flow and Von Neumann architecture using an array of processing elements, with compute and data movement designed to operate in parallel. It supports workloads including neural networks, computer vision, digital signal processing, and other parallel algorithms.
Quadric pairs the processor architecture with the Chimera SDK and compiler toolchain. Developers can import ONNX models from frameworks such as PyTorch and TensorFlow, quantize and compile models, write custom C++ kernels, simulate execution, and profile performance before committing a design to silicon.
Why Quadric Matters
Quadric addresses an important challenge in edge AI: AI models can evolve considerably faster than the chips designed to execute them. Fixed-function accelerators may depend on predefined operators, creating difficulties when newer model architectures introduce workloads the hardware was not designed to support.
Quadric takes a software-programmable approach to AI acceleration, allowing new operators and custom workloads to execute on the same processor architecture. This is particularly relevant for products such as vehicles, industrial systems, cameras, and robots that may remain deployed for years while their AI software continues to change. Its approach reflects the broader move toward more flexible on-device AI computing rather than relying exclusively on cloud-based inference.
Quadric Products & Platforms
Chimera GPNPU
Chimera is Quadric’s programmable neural processing unit IP for integration into SoCs. It is designed to execute AI inference as well as C++-based processing on a common architecture, reducing the need to divide workloads among separate accelerator, DSP, and processor blocks. Chimera can be configured across different performance and power requirements for applications ranging from low-power devices to larger autonomous systems.
Chimera SDK
The Chimera SDK is the software development and compilation environment for Chimera processors. It includes the Chimera Graph Compiler, Compute Library APIs, an LLVM C++ compiler, and an instruction-set simulator. Developers can use the SDK to compile neural-network graphs, create custom kernels, simulate workloads, and profile performance for different Chimera configurations.
Quadric DevStudio
DevStudio provides an environment for developers and SoC designers to evaluate AI models against Chimera hardware configurations. Users can work with existing models or upload their own, compile them through the Chimera toolchain, and examine performance before physical hardware is available.
Quadric Applications
Quadric’s technology targets AI inference that needs to run locally inside devices rather than exclusively in cloud data centers. Automotive applications include ADAS perception, sensor fusion, and driver monitoring, while industrial applications include machine vision, defect detection, and predictive maintenance.
The Chimera architecture can also support robotics and autonomous systems, where processors must handle navigation, obstacle detection, perception, and other real-time workloads. Additional applications identified by Quadric include smart cameras and surveillance, consumer wearables, networking equipment, and other power-constrained edge devices requiring local AI processing.
Trending Companies
Frequently Asked Questions
What is a GPNPU?
GPNPU stands for General-Purpose Neural Processing Unit. Quadric uses the term for its programmable processor architecture, which combines neural-network acceleration with the ability to execute general data-parallel algorithms and custom C++ code.
How is Quadric's Chimera different from a traditional NPU?
Traditional NPUs commonly accelerate a predefined collection of neural-network operations. Chimera is designed to be programmable, allowing developers to implement additional operators and custom kernels through software rather than relying exclusively on fixed-function hardware.
Does Quadric manufacture AI chips?
Quadric’s primary offering is licensable processor IP rather than finished consumer processors. The Chimera core is supplied as synthesizable RTL that semiconductor companies can integrate into their own SoCs.
Can Quadric Chimera run large language models?
Yes. Quadric’s current SDK documentation identifies support and demonstrations involving models and architectures including Llama-2, Qwen3-8B, and DeepSeek-R1-Distill, alongside vision and other AI workloads. Actual performance and model feasibility depend on the selected hardware configuration and workload.
What AI frameworks does the Chimera SDK support?
The Chimera toolchain can import models in ONNX format, including models originating from frameworks such as PyTorch and TensorFlow. Developers can then quantize, compile, simulate, validate, and profile models for Chimera processors.
Do You Have an Update for This Company?
Have an update, correction, or additional information about this company? Help us keep the futureTEKnow company database up to date by submitting the latest company information.
Related Companies
Latest Articles
Are your AI projects stuck on your own data?
AI projects often stall because business data isn’t consistent or trustworthy. Here’s how to identify data gaps before they become
X Square Robot’s $276M Bet on Embodied AI and China’s General-Purpose Robots
X Square Robot has raised $276M from Xiaomi, Sequoia China, and other internet giants to scale its WALL-A embodied AI
EVAS Intelligence’s RISC-V AI Chips Land 1.5 Billion Yuan Series B to Power Global Expansion
EVAS Intelligence has raised 1.5 billion yuan to mass‑produce its RISC-V Epoch AI chips, deepen its full‑stack platform, and accelerate
Orkes Raises $60M to Orchestrate Enterprise AI Agents in Production
Orkes has raised 60 million dollars to turn its Netflix‑born workflow engine into a control plane for enterprise AI agents.
Sillage’s AI signal engine raises €1.7M to help enterprise sales teams follow the right buying signals
Paris-based Sillage has raised €1.7 million to launch an AI signal engine that helps enterprise sales teams follow the right
Cloneable’s agentic AI platform targets the infrastructure knowledge gap with $4.6M seed round
Cloneable is launching an agentic AI platform for infrastructure operations that captures institutional knowledge from retiring experts and turns it
Algorized
Algorized is a deep-tech company developing edge-AI models that give machines real-time awareness of people using wireless sensors such as
Reliable Robotics Secures $160M to Scale Autonomous Aircraft Systems for Commercial Deployment
Reliable Robotics has secured $160M to scale production and deployment of its Reliable Autonomy System. This funding marks a pivotal
Ricursive Superintelligence’s $500M bet on self-improving AI raises the stakes in the AGI funding race
Excerpt: Ricursive Superintelligence has raised at least $500 million to build self‑improving AI, with GV and Nvidia backing a four‑month‑old
Bond AI raises US$2M to automate SME accounting in Brazil’s brutal tax maze
Brazilian startup BOND has raised US$2M to automate accounting for SMEs in Brazil’s complex tax system. Combining AI with human
Loop’s AI supply chain platform raises $95M Series C to predict disruptions before they hit
Loop just raised a $95M Series C to expand its AI-native supply chain platform, turning messy logistics data into early
London AI startup Lua wants to give every business its own AI workforce with $5.8M seed round
Linkedin X-twitter-square Facebook-square Startups & Business News AI agents are finally moving from demos to the day-to-day stack of real
futureTEKnow is focused on identifying and promoting creators, disruptors and innovators, and serving as a vital resource for those interested in the latest advancements in technology.
© 2026 All Rights Reserved.