Ollama Worker Keeps Redownloading Model
Hi, I’m trying to set up my first serverless Ollama (0.9.6) endpoint with an attached network volume (85 GB) to store the downloaded model (74 GB). Unfortunately, once the model has been fully downloaded, the worker starts downloading it again, and continues to do so indefinitely.
Is there something I might be overlooking?
2 Replies
Unknown User•4mo ago
Message Not Public
Sign In & Join Server To View
What we did was set OLLAMA_MODELS to point to a mounted network drive ...basically create a network drive then launch a pod with it attached and ssh in and install ollama and the model. i'm guessing you could do the same with docker. check your env var for OLLAMA_MODELS