# NeveAI

A local AI orchestration platform for running large language models offline, with a FastAPI backend and a SvelteKit interface for chat, document work and task automation.

NeveAI is a local AI ecosystem designed to run large language models offline, without a subscription or external API. It lets a user chat with models, work with documents, automate tasks and generate images, with everything processed on the user's own computer.

The system combines an asynchronous FastAPI backend with a SvelteKit 5 frontend, an SQLite database managed through SQLAlchemy and Alembic, and the llama.cpp engine for GGUF models, with GPU acceleration through CUDA or Vulkan. An installer detects the GPU, downloads a matching llama.cpp build, creates a Python virtual environment, installs dependencies, builds the frontend and creates the needed folders and configuration.

It targets Windows as its primary platform, with Linux and macOS support requiring manual adaptation.

Website: https://github.com/Etamus/NeveAI