Open-Assistant
Open-Assistant copied to clipboard
Local Inference does not work
Trying to get local inference to work, this is the docker compose command I am running:
docker compose --profile ci --profile inference up --build --attach-dependencies
I have also tried different permutations with the frontend-dev
and backend-dev
profiles. They also do not work.
It seems like the chat request gets put onto a queue, but the inference worker perhaps never pulls from it? Here is an image: