chenhunghan / ialacol

🪶 Lightweight OpenAI drop-in replacement for Kubernetes
MIT License
143 stars 17 forks source link

Downloading models fail with timeouts, retry is not enabled. #58

Closed DavidARivkin closed 1 year ago

DavidARivkin commented 1 year ago

When downloading a number of models (clearly from Hugging Face), I get the following error every time, seemingly at exactly the same point as well for each respective model that I have tried, even very small ones. ERROR: Error downloading model: HTTPSConnectionPool(host='cdn-lfs.huggingface.co', port=443): Read timed out.

I think the code needs to be modified to include retries.

It would also be helpful if the documentation described how we can download models manually from within the pod.

chenhunghan commented 1 year ago

Hello, you can add DOWNLOAD_DEFAULT_MODEL: False to skip the auto download, and manually download to /app/models in the pod, you are right about the documentation, will be update when I have time.

DavidARivkin commented 1 year ago

Since the pod does not have curl, wget, etc. how do you suggest we manually download the file?

3deep5me commented 1 year ago

Maybe an init container which downloads the models is a option.