ST Edge AI Suite

Dive into the edge AI world

Find out more about the products and tools that enable edge AI in your embedded project.

View all

STM32 Summit | Tech dive

Explore computer vision use cases

Hear from our ST Partners on how they used the STM32N6 MCU with AI acceleration in their applications.

Vincent Richard

AI solutions Product Marketing Manager

Watch the keynote

All the tools available in the ST Edge AI Suite

X-LINUX-AI

HAND POSTURE TOF

HIGH SPEED DATALOG

MEMS STUDIO

NANOEDGE AI STUDIO

ST EDGE AI CORE

ST EDGE AI DEVELOPER CLOUD

MODEL ZOO

STELLAR STUDIO

STM32CUBE.AI

ST AIoT CRAFT

See all tools

All frameworks compatible with ST edge AI solutions

All authorized partners

The ST Edge AI Suite is supported by a growing ecosystem of partners, expanding the available resources and expertise for your project.

See all our partners


Architecture	Core Arm^® Cortex^®-M55	Operating frequency 800 MHz	CoreMark 3.360	RAM 4.2 Mbytes contiguous RAM
Edge AI and multimedia	ST Neural-ART Accelerator 600 GOPS	Computer vision pipeline Parallel and MIPI CSI-2 camera I/F Image signal processor (ISP)	Graphic accelerators Chrom-ART: 2D GPU Chrom-GRC for non-square displays NeoChrom: 2.5D GPU	Multimedia accelerators H.264 encoder 1080p15 and 720p30 JPEG encoder and decoder
Security	Target certifications SESIP Assurance Level 3 PSA Certified Level 3	TrustZone^® YES	Resource isolation framework YES	Secure secret provisioning YES

Go to STM32N6 product page

FAQ

Quickly find clear, concise answers to the most common questions.

Frequently asked questions on how to deploy edge AI in embedded projects.

Edge AI refers to the enabling technology that runs AI algorithms and models directly on devices such as microcontrollers, microprocessors, and sensors embedded in industrial and automotive applications. Deploying AI at the edge enables real-time data processing directly at the source of data collection, offering faster response times, enhanced data security, and greater bandwidth efficiency. Read more about the advantages it offers here.

For all ST devices: the ST Edge AI Core CLI version and the ST Edge AI Developer Cloud allow users to optimize and evaluate AI model performance on any ST hardware.
For STM32 MCUs: STM32Cube.AI (X-CUBE-AI) enables neural network optimization. NanoEdge AI Studio is an AutoML tool.
For STM32 MPUs: developers can use AI for OpenSTLinux (X-LINUX-AI) and the STM32MP2 offline compiler for Linux AI frameworks.
For Stellar MCUs: StellarStudioAI is a software package for neural network optimization and deployment.
For MEMS sensors with a machine learning core: the online tool ST AIoT Craft and the desktop tool MEMS Studio can be used for data analysis, algorithm design, and model optimization. The MLC model zoo provides pre-optimized models.
For MEMS sensors with an ISPU: the MEMS Studio enables data analysis and model optimization. The ISPU model zoo provides pre-optimized models."

The ST Edge AI Suite facilitates the deployment of AI models by allowing users to easily find the right tool for their project:

Data logging: capturing the sensor data necessary for AI model training.
Auto ML: automatically generating optimized machine learning algorithms.
Model optimization: optimizing AI models and generating associated code for target devices.
Validation and testing: ensuring model performance meets deployment criteria.
Online benchmarking: testing model performance on ST hardware using the cloud.

Embedded developers can also benefit from:

The model zoo: simplifying the deployment of AI models on supported devices.
Documentation for more guidance through the deployment process.

The ST Edge AI Suite is compatible with a wide range of sensors as shown in the breakdown:

Time series sensors: accelerometers, gyroscopes, magnetometers, temperature sensors, ToF ranging sensors and other sensors that output data over time.
Audio sensors: microphones are the primary sensors for capturing audio data.
Vision sensors: cameras (RGB, B&W, IR), Time of Light sensors, radar, lidar, and more.

The ST Edge AI Suite is optimized for ST sensors, including MEMS devices with an MLC and the ISPU. It supports any sensor as long as the data provided is compatible with the tool requirements.

The tools in the ST Edge AI Suite can support different types of data:

High Speed Datalog:
- Time Series Data

NanoEdge AI Studio:
- Time Series Data

STM32Cube.AI (X-CUBE-AI):
- Time Series Data
- Audio Data
- Vision Data

MEMS Studio:
- Time Series Data (from the MEMS sensor)

StellarStudioAI:
- Time Series Data
- Audio Data

AI for OpenSTLinux (X-LINUX-AI):
- Time Series Data
- Audio Data
- Vision Data

ST AIoT Craft:
- Time Series Data (from the MEMS sensor)

Visit the edge AI Community

Your stepping stone to enabling edge AI on MCUs, MPUs, and smart sensors.

Dive into the edge AI world

Explore our edge AI lab

Generate optimized ML libraries

Deploy in-sensor AI with an online tool

Benchmark AI neural network models

Program and evaluate MEMS sensors

Demo: classify motor faults

STM32 Summit | Tech dive

Explore computer vision use cases

All the tools available in the ST Edge AI Suite

X-LINUX-AI

HAND POSTURE TOF

HIGH SPEED DATALOG

MEMS STUDIO

NANOEDGE AI STUDIO

ST EDGE AI CORE

ST EDGE AI DEVELOPER CLOUD

MODEL ZOO

STELLAR STUDIO

STM32CUBE.AI

ST AIoT CRAFT

"NanoEdge AI Studio allowed us to create the E-SmellAir with remarkable speed and efficiency. Its intuitive interface let us build and deploy customized gas classification models in record time."

All frameworks compatible with ST edge AI solutions

All authorized partners

STM32 microcontrollers with AI acceleration

Feature overview

FAQ

Frequently asked questions on how to deploy edge AI in embedded projects.

What is edge AI?

Which tool(s) can I use according to my target device?

How does the ST Edge AI Suite support the deployment of AI models on hardware devices?

Are there any costs associated with using the ST Edge AI Suite?

What is the ST Edge AI Suite?

Can I use any type of sensor with the tools provided in the ST Edge AI Suite?

What types of data are supported by each tool?