NeuroMem® is a technology of General Vision

Unleash the Intel's Curie Neurons of your 101

Train on board, not remotely • Identify similarities • Detect anomalies • Deep Learning

Learn More

NeuroMem for
Next Gen Wearables

Learn normal behavior & conditions • Classify via context awareness • Detect with multiple sensors

Learn More

NeuroMem for
Next Gen Cameras, Drones, & Robots

Identify people and objects • Adaptive learning as target changes • Find new and different events
Learn More

NeuroMem for
Next Gen Industrial Sensors

Learn, classify & detect anomalies • Adapts to production changes • Reduce human inspection • Improve quality throughput
Learn More

NeuroMem is a Unique Pattern-Recognition Technology

It's Adaptive

Learn on the chip thanks a build-in model generator  identifying novelty autonomously and discarding duplicates


It's Real Time

Recognize patterns in micro-seconds regardless of the number of reference models stored in the neurons.

It's Energy Efficient

The NeuroMem neurons are capable of delivering 100,000 recognitions per second and within milliWatts

It's Available NOW

The NeuroMem neurons are available on  ASICs, SOCs, and FPGAs and can be evaluated on a variety of platforms

GV Company News

New CurieNeurons Kit from Maker Collider

Maker Collider is introducing its CurieNeurons Kit, an easy-to-use tool for makers of all levels who want to have a shot at building artificial intelligence related IoT projects. It enables users to build smart hardware capable of pattern learning and classification... read more

Collaboration with Mando Hella Electronics

Mando-Hella Electronics and General Vision started a collaboration to develop applications targeted at Advanced Driver Assistance and Monitoring and relying on the NeuroMem neuromorphic technology to deploy powerful image recognition and multi-sensor pattern... read more

Collaboration with Nepes

Nepes and General Vision started a collaboration to design and manufacture a new line of NeuroMem neuromorphic chips ranging from 500 neurons to several thousands neurons per chip. It is anticipated that the collaboration will go beyond dedicated NeuroMem chips and... read more



Design IoT applications and smart non-connected appliances integrating autonomous pattern learning and recognition with 1024 silicon neurons. Easy to program through the Arduino IDE, Eclipse or else, the BrainCard can be interfaced to a wealth of sensors and actuators.


Develop applications which can learn and classify patterns at high speed whether they come from text, network packets, images, audio signal, biosensors, and more. The expandable network of 4096 silicon neurons can receive patterns from the USB port or directly from the reconfigurable on-board FPGA.