This hands-on workshop empowers participants to take control of their AI experience by deploying language models locally on their devices. Using accessible tools like LM Studio ( https://lmstudio.ai/ ) and Fullmoon, participants will learn to implement distilled DeepSeek models and other optimized models. The workshop focuses on the practical aspects of local deployment, including enhanced privacy, offline accessibility, and customization options. Participants will gain practical experience running AI models on their hardware through guided hands-on exercises. This workshop emphasizes theoretical understanding and practical implementation, ensuring participants leave with a functioning local LLM setup and comprehensive knowledge of its applications and limitations.

Learning Outcomes:

After completing the session, learners will be able to

1. Understand why local deployment enhances privacy, security, offline access, and customization. – Benefits of local LLM deployment
2. Install and Configure LM Studio and Fullmoon with specific DeepSeek models. – Setup process mastery
3. Assess the practical limitations of locally deployed models, including tradeoffs between model size, response quality, and hardware constraints. – Practical application evaluation

Required Hardware Specifications (Optional):

• PC: 6GB+ VRAM
• Mac: M1 or better with 8GB+ RAM
• Mobile: iPhone/iPad with A17/M1 processor or better
• Latest operating system required
• Devices must be ready for in-session software installation