[Question]: Does ragflow use image data when doing retrieval?

[aaronchen2k]

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

@dosu

Does ragflow use image data when doing retrieval? I will decide whether to set the image_context_size parameter to 0 based on this.

[dosubot]

RAGFlow does not use image data or image embeddings during retrieval—retrieval is always text-based, even for documents that contain images. Images are processed during parsing (using OCR or vision-language models) to extract text, and only this text is used for retrieval. The image_context_size parameter controls how much image context is attached to parsed content, but it does not affect retrieval itself. If you want to exclude image context from retrieval, set image_context_size to 0 in your parser_config. For more details, see this explanation and this config guidance.

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}

   

[KevinHuSh]

RAGFlow does not use image data or image embeddings during retrieval.