Skip to content
Scan a barcode
Scan
Paperback LOCAL AI ON NVIDIA GPUs CUDA Projects Handbook: Build, Accelerate, Optimize, and Deploy Local LLMs with CUDA, PyTorch, TensorRT-LLM, Ollama, llama.cpp Book

ISBN: B0HMFDJNTY

ISBN13: 9798160896748

LOCAL AI ON NVIDIA GPUs CUDA Projects Handbook: Build, Accelerate, Optimize, and Deploy Local LLMs with CUDA, PyTorch, TensorRT-LLM, Ollama, llama.cpp

LOCAL AI ON NVIDIA GPUs
CUDA Projects Handbook
Turn your NVIDIA GPU into a high-performance local AI platform.
Running a local LLM is easy. Engineering one that uses CUDA efficiently, manages VRAM intelligently, delivers consistent throughput, and scales reliably is a different challenge.
Local AI on NVIDIA GPUs: CUDA Projects Handbook is a practical, project-driven guide to building, accelerating, profiling, optimizing, and deploying local AI workloads on NVIDIA hardware. It covers the complete path from GPU and CUDA fundamentals to production-ready inference with PyTorch, TensorRT-LLM, Ollama, llama.cpp, vLLM, quantization, RAG, multimodal AI, and multi-GPU systems. LOCAL AI ON NVIDIA GPUs CUDA Projects Handbook.docx
Instead of treating performance as guesswork, you will build CudaForge AI, an evolving local AI system where every optimization is measured, compared, and verified. You will learn to distinguish driver, CUDA runtime, toolkit, framework, and inference-runtime problems; benchmark real workloads; identify bottlenecks; and make evidence-based GPU engineering decisions. LOCAL AI ON NVIDIA GPUs CUDA Projects Handbook.docx
Inside, you will learn how to:
Configure a reliable NVIDIA CUDA development environment
Measure TTFT, prefill and decode speed, throughput, VRAM, utilization, power, and temperature
Run and optimize local LLMs with PyTorch and CUDA
Use Tensor Cores, precision modes, quantization, batching, and KV-cache optimization
Profile workloads with PyTorch, NVIDIA Nsight Systems, and Nsight Compute
Compare Ollama, llama.cpp, vLLM, and TensorRT-LLM
Build GPU-accelerated RAG, embeddings, and multimodal applications
Scale inference across multiple GPUs
Containerize, monitor, troubleshoot, and operate a reproducible local AI service
The book progresses through ten practical CudaForge projects, from a verified GPU workstation and benchmarking harness to a complete, observable, multi-GPU AI platform. LOCAL AI ON NVIDIA GPUs CUDA Projects Handbook.docx
Whether you are a developer, AI/ML engineer, systems engineer, technical learner, or NVIDIA GPU enthusiast, this handbook gives you the practical skills to move beyond simply running models-and start engineering local AI systems that are faster, measurable, reproducible, and ready for real workloads.

Recommended

Format: Paperback

Condition: New

$35.24
Save $0.74!
List Price $35.98
Ships within 2-3 days
Save to List

Customer Reviews

0 rating
Copyright © 2026 Thriftbooks.com Terms of Use | Privacy Policy | Do Not Sell/Share My Personal Information | Cookie Notice | Cookie Preferences | Accessibility Statement
ThriftBooks ® and the ThriftBooks ® logo are registered trademarks of Thrift Books Global, LLC
GoDaddy Verified and Secured