9798184784557 - mastering ollama: run local llms on your own hardware, eliminate cloud ai costs, and ship 5 real projects in 30 days di stallard, smith (5 risultati)
- Brossura
Da: PBShop.store US, Wood Dale, IL, U.S.A.PBShop.store US
Contatta il venditoreVenditore con 5 stelleCondizione: Nuovo
EUR 19,91
Spedizione gratuitaSpedito in U.S.A.Quantità: Più di 20 disponibili
PAP. Condizione: New. New Book. Shipped from UK. Established seller since 2000.
- Brossura
Da: PBShop.store UK, Fairford, GLOS, Regno UnitoPBShop.store UK
Contatta il venditoreVenditore con 5 stelleCondizione: Nuovo
EUR 18,52
EUR 4,84 spedizioneSpedito da Regno Unito a U.S.A.Quantità: Più di 20 disponibili
PAP. Condizione: New. New Book. Shipped from UK. Established seller since 2000.
- Brossura
Da: AHA-BUCH GmbH, Einbeck, GermaniaAHA-BUCH GmbH
Contatta il venditoreVenditore con 5 stelleCondizione: Nuovo
EUR 23,95
EUR 30,50 spedizioneSpedito da Germania a U.S.A.Quantità: 2 disponibili
Taschenbuch. Condizione: Neu. Neuware.
- Brossura
- Print on Demand
Da: California Books, Miami, FL, U.S.A.California Books
Contatta il venditoreVenditore con 5 stelleCondizione: Nuovo
EUR 17,74
Spedizione gratuitaSpedito in U.S.A.Quantità: Più di 20 disponibili
Condizione: New. Print on Demand.
- Brossura
- Print on Demand
Da: CitiRetail, Stevenage, Regno UnitoCitiRetail
Contatta il venditoreVenditore con 5 stelleCondizione: Nuovo
EUR 22,16
EUR 43,06 spedizioneSpedito da Regno Unito a U.S.A.Quantità: 1 disponibili
Paperback. Condizione: new. Paperback. You are paying for AI you do not own, running on servers you cannot see, processing data you were never supposed to share.Last month a developer pasted a database schema into ChatGPT to debug a query. Three weeks later his company's legal team was still writing incident reports. He is not a…n outlier. He is the rule.The second problem is money. OpenAI charges per token - every prompt, every completion, every retry when the API times out during your batch job. A small team lands at $200 to $265 per month before anyone accounts for context overhead and retry logic. That number doubles when the team grows. There is no ceiling.The third problem is the one nobody admits. You tried running a model locally. You spent a Saturday fighting CUDA errors, ended up at 3 tokens per second, and went back to the API. Not because local AI does not work. Because nobody showed you how to make it work on your hardware, with the right model, configured correctly from the start.This book fixes all three. This weekend.By the time you finish the last chapter you will have shipped five working AI systems running on your own hardware, consuming zero API credits, leaking zero data, and costing nothing beyond the electricity to run them.Project 1 - Local Coding Assistant (Days 1-6)The GitHub Copilot replacement developers have been looking for. Private, fast, running inside VS Code, and free after setup. Your code never leaves your machine.Project 2 - Private Knowledge Base with RAG (Days 7-12)Feed it your documents. Query them in plain English. Every answer cites its source. Your files never touch a server you do not own.Project 3 - Automated Document Summarizer (Days 13-18)A CLI pipeline that batch-processes folders of PDFs into structured summaries. No rate limits. No per-token billing. Runs overnight on a hundred documents.Project 4 - Local AI Chatbot with Web Interface (Days 19-24)A full web UI your entire team accesses from any browser on your network. Multi-model switching. Fully air-gapped. The internal AI tool your legal team will actually approve.Project 5 - Automation Agent with Tool Calling (Days 25-30)A reasoning agent that reads files, makes decisions, and writes output with no cloud API anywhere in the architecture. A fully autonomous local AI system. Working before the month ends.Before the projects you get the foundation most guides skip - the honest hardware truth, VRAM math, model selection by task, and the chapter this book refused to leave out: why your local setup is slow right now and the ten-minute benchmark that fixes it.Chapter 7 runs the full cost math. A used RTX 3060 at $220 pays for itself in under 60 days at $265 monthly API spend. Over 24 months the savings exceed $6,000 for a single developer.Chapter 8 tells you what local AI still cannot do - because the most expensive mistake you can make is deploying local inference into a workload it cannot handle. You will know the edge before you hit it.The models available in 2026 - Llama, Mistral, Qwen, Gemma - are production-grade. The hardware caught up. The gap between local and frontier closed faster than anyone expected.You are one book away from owning your AI stack completely. The only mistake is not buying it. This item is printed on demand. Shipping may be from our UK warehouse or from our Australian or US warehouses, depending on stock availability.
