What is the error for ./entrypoint.sh: line 21: 21 Killed uvicorn webservice_starlette:app --host 0.0.0.0 --port ${PORT}?

Inference Endpoints on the Hub
1

I’m currently encountering an error while computing the image generation on my deployed Inference Endpoints. The error message in the logs is as ./entrypoint.sh: line 21: 21 Killed uvicorn webservice_starlette:app --host 0.0.0.0 --port ${PORT}

I am using GPU · Nvidia L40S · 1x GPU · 48 GB but this piepline ran well on the NVIDIA GeForce RTX 4090 24GB.

Is there a way to fix this error?

The full logs can be seen through below link

https://drive.google.com/file/d/1p0u3v-6KVva2NgtBZfJKPCMgkiM-bqOW/view?usp=sharing

1 Like
2

Is it memory-related issue?

1 Like
3

I think it is because of using fuse_lora() method that is currently experimental API

1 Like
4

After removing the method, I got no error

1 Like