Ambiq debuts AI tools to cut power and speed up edge inference
Ambiq launched two new edge AI runtime solutions, HeliosRT and HeliosAOT, optimized for their Apollo SoCs to enhance AI performance and energy efficiency in edge computing.
HeliosRT is a power-optimized version of LiteRT (TensorFlow Lite for Microcontrollers) offering up to 3x improvements in inference speed and power efficiency. HeliosAOT is an ahead-of-time compiler that converts TensorFlow Lite models into embedded C code, reducing memory usage by 15–50% and improving deployment flexibility.
Both solutions address challenges in deploying AI on ultra-low-power devices like wearables, IoT sensors, and industrial monitors. Built on Ambiq’s patented SPOT technology, these tools deliver significant power consumption improvements for edge AI applications.
“The intersection of developer experience and power efficiency is our north star,” says Carlos Morales, VP of AI at Ambiq. “HeliosRT and HeliosAOT are designed to integrate seamlessly with existing AI development pipelines while delivering the performance and efficiency gains that edge applications demand. We believe this is a major step forward in making sophisticated AI truly ubiquitous.”
HeliosRT is available in beta, with a general release expected in Q3 2025, while HeliosAOT is in technical preview for select partners, with wider availability planned for Q4 2025.
The tools integrate seamlessly with existing AI development workflows and are supported by documentation, examples, and engineering assistance.
Article Topics
AI/ML | Ambiq | edge AI | Edge Inference | IoT
Comments