Addons
Ollama
The easiest way to get up and running with large language models locally
Models are stored in /config/ollama.
Utilize the OLLAMA_HOST="http://homeassistant.local:11434" environment variable to use Ollama CLI on another machine.
Refer to the Ollama documentation for further details..
Example for pulling models:
Using Ollama CLI:
export OLLAMA_HOST="http://homeassistant.local:11434"
ollama pull nomic-text-embed
Using curl:
curl http://homeassistant.local:11434/api/pull -d '{
"name": "nomic-text-embed"
}'
Ava Server will automatically pull models on start when they are requested from Ollama.