Intel ARC machine learning requests failed
Hi yall, I've been trying to troubleshoot this for a while, but I just cant seem to get machine learning working with my Intel a380 on unraid.
I've installed immich with the recommended compose method. What's weird is that immich does seem to be aware of the card and temporarily starts attempting to use it before it errors out (see video with machine learning log, server log, and intelgpu top). I have tried changing around the ML address in the app and also setting IMMICH_HOST to 0.0.0.0 but neither of those methods has worked. I've attached my compose and ML files, the env file is currently set to default.
I've installed immich with the recommended compose method. What's weird is that immich does seem to be aware of the card and temporarily starts attempting to use it before it errors out (see video with machine learning log, server log, and intelgpu top). I have tried changing around the ML address in the app and also setting IMMICH_HOST to 0.0.0.0 but neither of those methods has worked. I've attached my compose and ML files, the env file is currently set to default.
docker-compose.yml2.75KB
hwaccel.ml1.38KB
