The Aspinity
AML200
The only edge AI processor to deliver TOPS level performance in a µW power footprint
The AML200 analog neural processor supports up to 2 TOPS peak performance with typical applications running <100uW
AML200 Features
- Optimized for autonomous sensing and RF/analog imaging for battery-operated edge and IoT
- AML200 supports 125k parameters and the NPU IP is scalable beyond 125K
- Designed at 22nm for efficiency, RF support, and compatibility with mainstream processors
- Scalable and flexible analog neural network supports a broad range of model types
- Available as chip, die, and IP
- Sampling in Q1 2025
Efficiency Comparison of Mainstream AI Architectures
Applications
Aspinity AML100 integrates with any embedded system to achieve an order-of-magnitude reduction in system power consumption
