Local AI Books Bundle

This book is a complete, practical guide to running large language models entirely on your own Mac. If you have an M1, M2, or M3 MacBook and want to deploy open-source LLMs locally for privacy, cost savings, or independence from cloud APIs, this book takes you from zero to mastery. You will learn the hardware architecture that makes Apple Silicon uniquely suited for this work, master every major inference framework (llama.cpp, Ollama, MLX, and Hugging Face Transformers), understand quantization strategies that fit billion-parameter models into your unified memory, and build real applications with local AI. Every technique described is fully open-source, reproducible on macOS, and grounded in real benchmarks and working code.

This book teaches you how to run decision models, a distinct class of AI systems optimized for fast, structured judgments, entirely on your own hardware using the open-source Ollaya project. You will learn what Jev-style decision models are, how they differ from conventional large language models, and why that difference matters for building reliable software. From installation to production deployment, this guide walks you through installing Ollaya, pulling models, defining typed questions, calling the API from Python and JavaScript, integrating with AI agents via MCP, designing decision schemas that work in the real world, measuring performance, and operating a local decision service at scale. By the end, you will be able to take a problem that currently uses an LLM, cost money, and leaks data to the cloud, and replace it with a local model that answers in milliseconds with typed outputs your code can act on directly.

Strata is a purpose-built inference runtime that makes it possible to run the 125B-parameter Qwen3.8-Flash-Next model at interactive speed on a single gaming GPU with as little as 12 GB of VRAM. This book explains exactly how it works, why it works and how to make it work for you. You will learn the architecture of Qwen3.8-Flash-Next, the mechanics of extreme mixture-of-experts sparsity, the three-tier memory hierarchy that distributes the model across VRAM, system RAM and NVMe storage, the CPU/GPU heterogeneous execution model, the pipelined PCIe transfers that hide latency, the speculative decoding that multiplies throughput and the practical details of installation, configuration, performance tuning and troubleshooting. If you are an engineer, ML practitioner, or advanced hobbyist who wants to understand and deploy Strata locally, this book is your guide.

This book is a complete guide to building, deploying, and optimizing local large language model infrastructure for software development workflows. It covers everything from selecting the right GPU and assembling the workstation through serving quantized models, building RAG pipelines with vector databases, creating agentic coding assistants, and integrating local inference with Claude Code. Written for developers, researchers, and AI engineers who want full control over their AI stack. This is the practical reference you need to go from zero to a production-ready local AI development system.

This book teaches you how to build a complete, production-grade local search engine that indexes diverse document types, performs hybrid keyword and semantic retrieval, powers retrieval-augmented generation with local language models, and runs entirely on your own hardware. You will move from a minimal working prototype to a feature-rich system with OCR, multimodal embeddings, incremental indexing, caching, security, and crossplatform deployment. Every code example is complete, runnable Python built on current stable releases of open-source libraries including FastAPI, FAISS, Qdrant, Chroma, sentence-transformers, Ollama, and more.

This book teaches software engineers and ML practitioners how to select, fine-tune, evaluate, and deploy small language models as capable local AI agents. You will learn the foundations of transformer architectures and tokenization, master supervised fine-tuning and parameter-efficient methods like LoRA and QLoRA, build agentic capabilities for tool use and structured output, and ship production-ready systems on consumer hardware. Every chapter includes reproducible code, concrete configurations, and evidence-based comparisons so you can move from an unmodified base model to a specialized agent running entirely on your own machine.

A comprehensive systems-engineering handbook for building AI applications and autonomous agents that remain capable, private, controllable, secure, maintainable, reproducible, and operational even when the cloud is unavailable or intentionally absent. This book takes you from hardware selection through Linux configuration, inference optimization, RAG pipelines, agent architecture, security hardening, reliability engineering, and production operations with a complete reference implementation that evolves across every chapter. Designed for software engineers, AI/ML engineers, systems engineers, platform engineers, DevOps/SRE practitioners, security engineers, researchers, technical architects, and advanced developers who need to build real local-first AI systems they own end-to-end.

This book teaches engineers how to build reliable evaluation systems for small and locally deployed language models. You will learn to measure performance accurately, compare models across hardware and runtimes, interpret results without being misled by benchmarks, and make evidence-based deployment decisions for real-world workloads. The guide combines rigorous experimental methodology with example code so you can independently establish an evaluation laboratory and produce trustworthy, reproducible results.