[Bug]: Image Handling with litellm Proxy for qwen-vl-plus

[NEWbie0709]

What happened?

I’m experiencing an issue when using litellm proxy to communicate with the qwen-vl-plus model for multimodal interactions. When I send an image URL directly to qwen-vl-plus, it processes the image as expected. However, when I configure litellm to handle the request and forward it to the qwen-vl-plus model, the same request fails. This is my config.yaml and this is the code for testing the litellm proxy

Other than that, I also tried using the ollama/llava model. When using the litellm proxy to send the request, it returns an error saying that it can't open the image. But when I directly ask the model using ollama, it can answer it.

Relevant log output

No response

Twitter / LinkedIn details

No response