Ollama (Local)

Ollama is a free, open-source local model runtime. No API key needed — Youkore connects to the local server.

1. Download and install

Download and install from the link below.

→ Download Ollama

2. Pull a model

After installation, open a terminal and download a model:

ollama pull llama3.2

3. Server is running

The server starts automatically and listens on http://localhost:11434.

4. Connect Youkore

No key needed — Youkore connects automatically.

note

Ollama is free, open-source and works offline. Supports NVIDIA, AMD and Apple Silicon GPUs.