Mythic launches power-sipping AI chip

Edge-focused AI chip consumes one-tenth the power of a desktop GPU

Semiconductor startup Mythic has launched an artificial intelligence (AI) processor that it says will deliver the computing capabilities of a GPU for a tenth of the power.

The M1076 Mythic AMP chip targets edge computing AI applications with power and footprint limitations. Like other edge-based AI processors, the Mythic CPU runs a neural network to handle a process called inferencing. This takes new input, such as a video signal from a camera in a manufacturing plant, and runs it through a trained AI model designed to recognize specific things — like an object on a production line.

The difference lies in how the chip processes data in its neural network. It uses a combination of analog circuitry and in-memory computing to save power. Rather than storing its data in slower memory that the chip must access when needed, it uses multiple memory arrays that handle the computing themselves using their memory elements as tunable resistors. 

This allows the chip to compute neural network inputs and outputs as voltages and currents. The memory arrays function as nodes in the neural network graph and can operate in parallel.

The unique chip architecture enables the company to fit more AI computing horsepower into a single chip, it says. The M1076 delivers up to 25 trillion operations per second (TOPS) using a 3-watt power draw. It's available as a single chip, a PCIe M2 card for low-footprint applications, and a PCIe card with up to 16 chips. The latter can deliver 400 TOPS for just 75 watts of power.

Related Resource

The AI-powered supply chain

Better demand forecasting and operational excellence

An airplane flying over a city - The AI-powered supply chain - whitepaper from DataRobotDownload now

To run neural networks trained by popular AI packages such as PyTorch, developers will need to retrain them to run on the chip's Analog Compute Engine (ACE) and then compile them for the chip's architecture.

Mythic says the chip will work well in video processing applications, including object detection and depth estimation, making it suitable for industrial machine vision, surveillance cameras, and augmented reality applications.

The company raised $70 million in series C funding in May, led by BlackRock and Hewlett Packard Pathfinder, bringing its total funding to $165.2 million.

Featured Resources

Consumer choice and the payment experience

A software provider's guide to getting, growing, and keeping customers

Download now

Prevent fraud and phishing attacks with DMARC

How to use domain-based message authentication, reporting, and conformance for email security

Download now

Business in the new economy landscape

How we coped with 2020 and looking ahead to a brighter 2021

Download now

How to increase cyber resilience within your organisation

Cyber resilience for dummies

Download now

Most Popular

How to find RAM speed, size and type

How to find RAM speed, size and type

16 Jun 2021
EU plans to launch bloc-wide cyber task force
cyber attacks

EU plans to launch bloc-wide cyber task force

22 Jun 2021
What is HTTP error 400 and how do you fix it?
Network & Internet

What is HTTP error 400 and how do you fix it?

16 Jun 2021