Upgrade to Pro — share decks privately, control downloads, hide ads and more …

1907 IBM Granite LLM and XR Device Technology C...

1907 IBM Granite LLM and XR Device Technology Collaboration Live Demo and Knowledge Talk

This is my presentation in IBM TechXchange 2025 Orlando. Title is "1907 IBM Granite LLM and XR Device Technology Collaboration Live Demo and Knowledge Talk".

Thank you for graphic materials!

kage-design http://kage-design.com/
human pictogram 2.0 http://pictogram2.com
ICOOON MONO http://icooon-mono.com/
Adobe Stock https://stock.adobe.com/jp

Avatar for 1ft-seabass

1ft-seabass PRO

October 08, 2025
Tweet

More Decks by 1ft-seabass

Other Decks in Technology

Transcript

  1. Orlando, FL October 6–9 IBM TechXchange 2025
 Session code 1907


    Seigo Tanaka
 President, 1ft-seabass
 IBM Granite LLM and XR Device Technology Collaboration Live Demo and Knowledge Talk

  2. Introduction & Today's Architecture
 IBM TechXchange | © 2025 IBM

    Corporation 2
 Voice to IoT Control
 xR + Voice + AI + IoT

  3. Like this...
 
 "Turn on light" voice to "self-build LEGO

    IoT Light" IoT control demo video.
 
 IBM TechXchange | © 2025 IBM Corporation 3

  4. How It Works
 
 Detail : https://github.com/1ft-seabass/ibm-techxchange25-orlando/blob/main/01-how-it-works.md 
 IBM TechXchange

    | © 2025 IBM Corporation 4
 Ollama
 +
 Granite LLM
 Voice
 Node-RED
 IoT
 Local PC
 Illustration From Adobe Stock(Adobe Express)
 Meta Quest 3
 My Photo
 MCP
 AI

  5. Why Local try?
 
 IBM TechXchange | © 2025 IBM

    Corporation • I already made with IBM Code Engine(API part) and watsonx.ai (AI part) as Voice to IoT Control.
 • It's a nice and general cloud approach. But the data stays on public cloud.
 • IoT use cases tend to need local control such as Equipment control, Sensitive data handling, Network dependency risks.
 • So I wanted to try almost fully local setup with Local LLM! Granite + Ollama on local PC.
 
 5

  6. Previous : IBM Code Engine(API part) and watsonx.ai (AI part)


    
 Detail : https://github.com/1ft-seabass/ibm-techxchange25-orlando/blob/main/01-how-it-works.md 
 IBM TechXchange | © 2025 IBM Corporation 6
 watson.ai
 +
 High Tune Prompt
 Voice
 IoT
 Cloud
 Illustration From Adobe Stock(Adobe Express)
 Meta Quest 3
 My Photo
 MQTT
 IBM Code Engine
 AI
 API

  7. Today System(Local)
 
 Detail : https://github.com/1ft-seabass/ibm-techxchange25-orlando/blob/main/01-how-it-works.md 
 IBM TechXchange |

    © 2025 IBM Corporation 7
 Illustration From Adobe Stock(Adobe Express)
 Ollama
 +
 Granite LLM
 Voice
 Node-RED
 IoT
 Local PC
 Meta Quest 3
 My Photo
 MCP
 AI

  8. Why MCP?
 
 IBM TechXchange | © 2025 IBM Corporation

    • MCP is a nice bridge between human natural language command and common API calls.
 • Simpler than long and complex prompts for voice understanding and JSON schema rules.
 • MCP uses general programming languages. (TypeScript and Python...)
 • This code-based approach enables IoT control (MQTT protocol) Easy to integrate with existing systems!
 8

  9. Live Demonstration!
 
 Detail : https://github.com/1ft-seabass/ibm-techxchange25-orlando/blob/main/02-live-demonstration.md 
 
 IBM TechXchange

    | © 2025 IBM Corporation 9
 Ollama
 +
 Granite LLM
 Voice
 Node-RED
 IoT
 Local PC
 Meta Quest 3
 My Photo
 MCP
 AI
 Illustration From Adobe Stock(Adobe Express)

  10. Try and Error Points
 
 IBM TechXchange | © 2025

    IBM Corporation 10
 • This is "almost" local for now because Voice to Text part is public Whisper API (transcription).
 • Currently it has simple MCP client server call to MCP IoT control tool. I aim to adjust to AI agent approach.
 • Granite supports multi language response. But English response has good response speed.

  11. Key Insights
 
 
 IBM TechXchange | © 2025 IBM

    Corporation 11
 • Granite LLM's collaboration with Ollama levels up general API integration with IoT and XR through Node-RED.
 • Granite LLM's support for Function calling (structured output) is very powerful, making it easier to select tools for MCP-like IoT operations.
 • Ollama can connect with AI agent tools like Mastra. So it feels more collaboration with AI agent.

  12. More detail document...
 IBM TechXchange | © 2025 IBM Corporation

    12
 https://github.com/1ft-seabass/ibm-techxchange25-orlando 

  13. Conclusion
 IBM TechXchange | © 2025 IBM Corporation 13
 Granite

    LLM works great for prototyping in a closed environment with creative collaboration as well as in a cloud environment!

  14. Notices and disclaimers
 IBM, the IBM logo, and ibm.com are

    trademarks of International Business Machines Corporation, registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM trademarks is available on the Web at “Copyright and trademark information” at: www.ibm.com/legal/copytrade.shtml. 
 Certain comments made in this presentation may be characterized as forward looking under the Private Securities Litigation Reform Act of 1995. 
 Forward-looking statements are based on the company’s current assumptions regarding future business and financial performance. Those statements by their nature address matters that are uncertain to different degrees and involve a number of factors that could cause actual results to differ materially. Additional information concerning these factors is contained in the Company’s filings with the SEC. 
 Copies are available from the SEC, from the IBM website, or from IBM Investor Relations. 
 Any forward-looking statement made during this presentation speaks only as of the date on which it is made. The company assumes no obligation to update or revise any forward-looking statements except as required by law; these charts 
 and the associated remarks and comments are integrally related and are intended to be presented and understood together. 
 © 2025 International Business Machines Corporation. 
 All rights reserved. 
 This document is distributed “as is” without any warranty, either express or implied. In no event shall IBM be liable for any damage arising from the use of this information, including but not limited to, loss of data, business interruption, loss of profit or loss of opportunity.
 Case studies and client examples are presented as illustrations of how customers or IBM has used IBM products in production or test environments and the results they may have observed. Actual performance, cost, savings or other results in other operating environments may vary. 
 Workshops, sessions and associated materials may have been prepared by independent session speakers, and do not necessarily reflect the views of IBM. 
 Not all offerings are available in every country in which IBM operates. 
 Any statements regarding IBM’s future direction, intent or product plans are subject to change or withdrawal without notice. 
 IBM TechXchange | © 2025 IBM Corporation 15