Small Language Models for Mobile Devices (Paperback)

Thomas O. Greene

ISBN 13: 9798259071360

Published by Independently Published, 2026

About this Item

Paperback. Stop Renting Intelligence. Start Owning It. The Cloud is hitting a wall. Latency is killing your user experience. Privacy is becoming a legal minefield. And API costs are bleeding your startup dry.Now, the "God Models" have moved from massive data centers into the palm of your hand.In Small Language Models for Mobile Devices, visionary developer and engineer Thomas O. Greene reveals the blueprint for the most significant shift in computing since the smartphone itself: The Silicon Sovereignty.We are moving away from "Intelligence-as-a-Service" and toward "Intelligence-as-a-Utility." This book is your technical manifesto and hands-on guide to building, optimizing, and deploying high-performance AI that runs 100% offline, with sub-50ms latency, on standard Android and iOS hardware. What's Inside the Engine Room?The Architecture of Efficiency: Deep-dives into Phi-4, Gemma, and Llama-3-Mobile. Learn why "small" doesn't mean "weak" when you master Grouped-Query Attention (GQA) and Rotary Embeddings.The Magic of Quantization: Step-by-step techniques to squeeze 7B parameter models into 4GB of RAM using INT4, NF4, and the 1.58-bit Binary Frontier.Next-Gen Frameworks: Master ExecuTorch (PyTorch Edge), Apple MLX, and Android AICore to talk directly to the NPU silicon.Beyond Text: Deploy Multi-Modal SLMs that "see" through the camera and "hear" through the mic with native audio-to-audio processing.The Agentic Revolution: Build Large Action Models (LAMs) that navigate mobile UIs, booking rides and sending messages without a single cloud request.The Future is Liquid: An exclusive look at Liquid Neural Networks (LNNs)-the breakthrough for infinite context and constant memory footprints.Why This Book is Essential: Whether you are a Mobile Developer tired of "Cloud Fatigue," a Machine Learning Engineer fighting the "Memory Wall," or a Tech Leader demanding "Privacy-First" AI, this book provides the code, the math, and the strategy to win. The era of the "Frozen Snapshot" LLM is over. The era of the Fluid, Private, and Autonomous Mobile Agent has begun.Stop sending your users' data to a third-party server. Take the red pill of Data Sovereignty and build the private, powerful, and portable future today. This item is printed on demand. Shipping may be from multiple locations in the US or from the UK, depending on stock availability.

Seller Inventory # 9798259071360

Bibliographic Details

• Title: Small Language Models for Mobile Devices (Paperback)
• Publisher: Independently Published
• Publication Date: 2026
• Language: English
• ISBN 13: 9798259071360
• Binding: Paperback
• Condition: new

About this title

Synopsis

Stop Renting Intelligence. Start Owning It.

The Cloud is hitting a wall. Latency is killing your user experience. Privacy is becoming a legal minefield. And API costs are bleeding your startup dry.
Now, the "God Models" have moved from massive data centers into the palm of your hand.
In Small Language Models for Mobile Devices, visionary developer and engineer Thomas O. Greene reveals the blueprint for the most significant shift in computing since the smartphone itself: The Silicon Sovereignty.
We are moving away from "Intelligence-as-a-Service" and toward "Intelligence-as-a-Utility." This book is your technical manifesto and hands-on guide to building, optimizing, and deploying high-performance AI that runs 100% offline, with sub-50ms latency, on standard Android and iOS hardware.

What’s Inside the Engine Room?

• The Architecture of Efficiency: Deep-dives into Phi-4, Gemma, and Llama-3-Mobile. Learn why "small" doesn't mean "weak" when you master Grouped-Query Attention (GQA) and Rotary Embeddings.
• The Magic of Quantization: Step-by-step techniques to squeeze 7B parameter models into 4GB of RAM using INT4, NF4, and the 1.58-bit Binary Frontier.
• Next-Gen Frameworks: Master ExecuTorch (PyTorch Edge), Apple MLX, and Android AICore to talk directly to the NPU silicon.
• Beyond Text: Deploy Multi-Modal SLMs that "see" through the camera and "hear" through the mic with native audio-to-audio processing.
• The Agentic Revolution: Build Large Action Models (LAMs) that navigate mobile UIs, booking rides and sending messages without a single cloud request.
• The Future is Liquid: An exclusive look at Liquid Neural Networks (LNNs)—the breakthrough for infinite context and constant memory footprints.

Why This Book is Essential:
Whether you are a Mobile Developer tired of "Cloud Fatigue," a Machine Learning Engineer fighting the "Memory Wall," or a Tech Leader demanding "Privacy-First" AI, this book provides the code, the math, and the strategy to win.

The era of the "Frozen Snapshot" LLM is over. The era of the Fluid, Private, and Autonomous Mobile Agent has begun.
Stop sending your users' data to a third-party server. Take the red pill of Data Sovereignty and build the private, powerful, and portable future today.