inference

Practical guide to deploying local AI on Linux: pick suitable CPU/GPU, install drivers and CUDA, set up Python envs, install PyTorch/TensorFlow, fetch models (Hugging Face), and run with Ollama, ONNX, or ggml/gguf. Covers quantization, GPU acceleration, REST serving, containers, privacy, benchmarks, and fixes for common build, memory, and driver issues.