Texas Instruments introduced two new microcontroller families with edge artificial intelligence capabilities: the MSPM0G5187 and AM13Ex MCUs. Both integrate TI’s TinyEngine neural processing unit (NPU), a dedicated hardware accelerator that optimizes deep learning inference to reduce latency and improve energy efficiency at the edge.
The TinyEngine NPU enables AI models to run with up to 90 times lower latency and more than 120 times lower energy utilization per inference compared to similar MCUs without an accelerator.
The MSPM0G5187 is an Arm Cortex-M0+ MCU that brings edge AI to simpler, smaller, and more cost-effective applications. With local computation, the TinyEngine NPU executes neural network operations in parallel to the primary CPU, minimizing flash memory footprint, lowering latency by up to 90 times per inference, and reducing energy utilization by more than 120 times per inference. Priced under US$1 in 1,000-unit quantities, it provides an affordable option for resource-constrained, battery-powered, or portable devices.
The AM13Ex MCUs integrate a high-performance Arm Cortex-M33 core, TinyEngine NPU, and advanced real-time control architecture into a single chip. This enables simultaneous sophisticated motor control and AI features without external components, reducing bill-of-materials costs by up to 30%. Key capabilities include maintaining precise real-time control loops for up to four motors while the TinyEngine NPU runs adaptive control algorithms for load sensing and energy optimization, and an integrated trigonometric math accelerator that performs calculations 10 times faster than CORDIC implementations for more precise and responsive motor control.
Both families are supported by CCStudio Edge AI Studio, a free development environment for model selection, training, and deployment. It includes more than 60 models and application examples, with additional tasks and models planned. The CCStudio IDE incorporates generative AI features for code development, system configuration, and debugging using industry-standard agents and models paired with TI data.
TI’s embedded processing portfolio includes these new MCUs to enable edge AI across applications from real-time monitoring in wearable health monitors and home circuit breakers to physical AI in humanoid robots.
The innovations are featured at TI’s booth at embedded world 2026, March 10-12, in Nuremberg, Germany (Hall 3A, Booth No. 131), including demonstrations of faster development with AI, enhanced performance with edge AI, deployment across factories, buildings, and vehicles, and the partner ecosystem.
Amichai Ron, senior vice president, Embedded Processing and DLP Products at TI, stated that TI invented the digital signal processor nearly 50 years ago and is now integrating the TinyEngine NPU across its microcontroller portfolio, including general-purpose and high-performance real-time MCUs, to make edge AI accessible through software, tools, devices, and ecosystem.
Bob O’Donnell, President and Chief Analyst at TECHnalysis Research, noted that while focus has been on AI acceleration in larger SoCs, interesting applications of AI can occur inside smaller MCUs, making consumer devices more intelligent and industrial devices more efficient, with AI-leveraging development tools broadening access for engineers.
Production quantities of the MSPM0G5187 MCU are available for purchase now on TI.com. The AM13E23019 MCU is available in preproduction quantities. Additional package and memory variants will be released by the end of 2026. Multiple payment and shipping options are available.
