Local AI Stack | Notes for Linux

Ctrl K

4 results

CUDA Setup on Ubuntu 24.04

Verify CUDA availability, install the CUDA Toolkit compiler (nvcc), configure environment variables, and confirm GPU access from Python.

Local Ollama API on Linux

Install Ollama, run it as a systemd service, pull models, and call the local API from curl and Python.

Local Text to Speech with XTTS v2

Set up Coqui XTTS v2 locally on GPU and turn a text script into a single-voice narration MP3, with sentence chunking for the 250 character limit.

Qdrant RAG Vector Index Workflow

Docker-based workflow for using Qdrant as a vector database in a RAG system, including manifest export, database source export, embedding execution, production deployment, and safe reindex checks.