### Check for existing issues

- [x] I have searched the existing issues and checked that my issue is not a duplicate.

### What happened?

I run a VLM chat call with an image to a proxied ollama model.

I'm on container image: docker.litellm.ai/berriai/litellm-database:v1.83.14-stable

I was able to fix the issue by manually installing pillow in the container image using pip.

### Steps to Reproduce

1. Run chat request to an ollama model (e.g. gemma4) 2. Observe 500 error.

### Relevant log output

```shell `Request Failed Error Code: 500 Message: litellm.APIConnectionError: ollama image conversion failed please run `pip install Pillow` Traceback (most recent call last): File "/app/.venv/lib/python3.13/site-packages/litellm/llms/ollama/common_utils.py", line 28, in _convert_image from PIL import Image ModuleNotFoundError: No module named 'PIL' During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/app/.venv/lib/python3.13/site-packages/litellm/main.py", line 3988, in completion response = base_llm_http_handler.completion( model=model, ...<14 lines>... client=client, ) File "/app/.venv/lib/python3.13/site-packages/litellm/ll…