GrAI Matter Labs, also known as GML, is a French technology company that specializes in developing ultra-low latency and low power processing AI processors for edge devices.
It focuses on disrupting inference at the endpoint device, particularly in the areas of audio and video processing with very low latencies. By processing data at the source, GML's "Life-ready" AI solutions offer almost instant insights and transformations without the typical delays and costs associated with cloud servers.
GrAI Matter Labs introduced its first-generation chip, GrAI One in October 2019, heralding it as the world's first AI processor optimized for ultra-low latency and low power processing at the edge.
A second generation was produced solely for a project GrAI Matter worked on with the US government.
GrAI Matter's third-generation GrAI VIP chip uses event-based sensing and sparsity for efficient image data processing, offering low-latency, low-power computer vision. The third-gen chip features a FP16-capable neuron flow fabric, dual Arm Cortex M7 CPUs, and can handle MobileNetv1-SSD at 30fps for 184mW, providing significant efficiency gains compared to GPUs.
The company's technology has shown superior performance compared to competitors like NVIDIA's Jetson Nano, offering 10x better performance at over 10x lower power consumption for tasks like Resnet50.
Key customers and partnerships
GML has collaborated with multiple companies to create advanced AI solutions. These partnerships involve working with Tuito (January 2023) to improve speech across various sectors, StreamUnlimited (September 2022) to demonstrate real-time music/media separation AI, FRAMOS (October 2021) to develop AI for industrial vision, and EZ-WHEEL October 2021) to design an AI wheel drive for robots.
Funding and financials
In November 2020, the company raised USD 14 million in funding led by iBionext. As per the company, the fresh capital was to be used towards partnering with application specialists and integrators, thereby launching its first “GrAI full-stack AI system-on-chip” platform.
By using this site, you agree to allow SPEEDA Edge and our partners to use cookies for analytics and personalization. Visit our privacy policy for more information about our data collection practices.