Fine-Tuning LLMs for Mobile: Run Your Own AI Model on a Phone (Live Demo) | SaaviGenAI | Blithe

In this technical session, Lakshmi walks through the complete process of fine-tuning a Large Language Model (LLM) and running it directly on a mobile device.

The session combines clear conceptual explanations with a live demonstration, showing how a general-purpose model can be specialised using modern techniques like LoRA-based fine-tuning, quantisation, and edge deployment.

The walkthrough covers the full pipeline — from preparing training data to deploying a compact model that runs locally on a phone without relying on cloud APIs.

Key topics covered in this session include:

• Understanding what fine-tuning really changes inside a model
• Differences between prompting, RAG, and fine-tuning
• Using LoRA for efficient fine-tuning with minimal compute
• Preparing datasets for supervised fine-tuning (SFT)
• Converting models to GGUF format for edge deployment
• Running quantised models locally using llama.cpp / MLC / Ollama
• Deploying LLMs on mobile devices with no cloud dependency

The session demonstrates how a model can move from general-purpose capability to domain specialization, and how optimized models can run efficiently on consumer hardware such as smartphones.

This is a practical walkthrough for developers, AI engineers, and practitioners interested in edge AI, model optimisation, and private on-device inference.

Hosted by @SaaviGenAI LLP focuses on helping professionals and organizations build practical AI capability. We work on AI fundamentals, upskilling teams, and helping enterprises consult, secure, their AI systems and data.

Social media Partner Blith the dedicated platform for micro-creators and hobby organisers to transition from passion to profession. By automating the logistical burden of coordination, ticketing and payments, we empower our hosts to reclaim their creative bandwidth.

#llm #llmsecurity #llms #llmops #genai #genaisecurity #saavigenai #cybersecuritynews #finetune #promptengineering #promts #cybersecuritynews #cybersecurityjobs

Видео Fine-Tuning LLMs for Mobile: Run Your Own AI Model on a Phone (Live Demo) | SaaviGenAI | Blithe канала SaaviGenAI