Trouble Enabling Image Upload with Self-Hosted Qwen-2.5VL-7B in Onyx

[YasaswiniSireddy]

I’ve been using Onyx to chat with documents and recently integrated my self-hosted Qwen-2.5VL-7B model, which is running on vLLM’s OpenAI-compatible server. Everything works fine for processing PDFs and other document formats, but I’m running into issues when trying to upload images in the chatbot interface.

What’s Happening?
When I upload an image, I get an error from onyx/web/src/app/chat/ChatPage.tsx.

I tried adding qwen to MODEL_NAMES_SUPPORTING_IMAGE_INPUT in onyx/web/src/lib/llm/utils.ts, hoping that would enable image support.

But i am still getting error

Is adding qwen to MODEL_NAMES_SUPPORTING_IMAGE_INPUT the right approach?
Does Onyx support image inputs for self-hosted models running on vLLM, or do I need to configure something else?

I am also facing a runtime error:
RuntimeError: No litellm entry found for openai/Qwen/Qwen2-VL-7B-Instruct-AWQ

How can I avoid the "No litellm entry found" error?

Would really appreciate any guidance on this! Thanks in advance.