[Question]: Regarding image search

[myplxdm]

Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (Language Policy).
• Non-english title submitions will be closed directly ( 非英文标题的提交将会被直接关闭 ) (Language Policy).
• Please do not modify this template :) and fill in all the required fields.

Describe your problem

Can Ragflow enable the uploading of images during a chat for analysis and response, where the content of the image is not OCR-based, such as a portrait? If so, how should the image be passed to the API? Should it be the image's URL?

[KevinHuSh]

It needs image2text multi-module LLM which should be configured at system model settings.

[myplxdm]

@KevinHuSh Thank you for your reply. If I implement this function through the chat API of ragflow, do I need to call the image2text model first to get the text and then pass it to the chat API? If so, does ragflow provide an API to call the image2text model set in the configuration?

[KevinHuSh]

Not yet.