The cutting edge of Edge AI - Locally generated AI (LLM) running on embedded processors

■Recommended points:
It's not just the cloud! Learn how generative AI (LLM) can be used on embedded devices and how to achieve it.

■Seminar Contents:

Learn about industry trends and the latest trends in Edge AI, as well as basic knowledge of generative AI/LLM/RAG. What kind of generative AI (LLM) systems can be realized using NXP's processors, which are ideal for embedded devices?

We will explain the generative AI (LLM) development and evaluation environment with a demonstration based on the i.MX 95 application processor.

LLM: Large Language Model
RAG: Retrieval Augmented Generation (a method for creating databases of domain-specific information)

■Target:

Those who are investigating how large-scale generative AI (LLM) can be achieved with embedded processors and how it can be used.

・ Those looking for something to add value to their embedded devices
Example: Ask questions by voice, in natural language, and get the right answer instead of reading instructions or manuals

・Those in charge of planning, semiconductor product selection, design, development, and manufacturing of embedded devices using Linux and processors

・ People who are testing Edge AI on PCs, NVIDIA Jetson, or Raspberry Pi, especially generative AI (LLM), but are concerned about long-term supply, reliability, power consumption, and price for practical use and mass production.

・ Those who want to implement AI (Edge AI) on embedded devices rather than on the cloud

・ Those who want to use generative AI (LLM) in environments with limited network connectivity

・ Those who want to manage the latest data and ensure security with RAG

・Those who want to catch up on the trending topic of generative AI (LLM)

■Registration:

・ Registration URL: Click here to apply

Date and time: Wednesday, June 18, 2025, 11:00 AM - 11:50 AM JST

Participation fee: Free

・Pre-registration required: Participation URL will be sent within 1 business day after registration.

・ Registration closes: June 17th (Tue) 17:00 (Please note that if you register after the registration closes, we may not be able to send you the participation URL.)

* Using the GoToWebinar platform Webcast mode. No application download is required and you can watch the event in your browser.

*In rare cases, you may not be able to watch depending on your environment. If you are using GoToWebinar for the first time, we recommend that you perform a pre-connection test using the "Check system requirements" link in the email containing the participation URL. Also, if you have any trouble while watching, try reloading your browser.

*Please note that we will not accept participation from trading companies related to competitors, or individuals for personal use (hobby users, or those who register with free email addresses such as Gmail/personal addresses).