[GH-ISSUE #13807] issue: #32569

New Issue

GiteaMirror · 2026-04-25T06:30:06-05:00

GiteaMirror commented

2026-04-25 06:30:06 -05:00

Originally created by @Urvesh71 on GitHub (May 12, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/13807

Check Existing Issues

I have searched the existing issues and discussions.
I am using the latest version of Open WebUI.

Installation Method

Docker

Open WebUI Version

0.6.7

Ollama Version (if applicable)

0.6.5

Operating System

Ubuntu

Browser (if applicable)

No response

Confirmation

I have read and followed all instructions in README.md.
I am using the latest version of both Open WebUI and Ollama.
I have included the browser console logs.
I have included the Docker container logs.
I have listed steps to reproduce the bug in detail.

Expected Behavior

While I do perform RAG with Document in Open WebUI, I am not getting a correct answers. I do not know what is wrong with it. When I upload the PDF and ask the question, everyatime it answers that there is not content provided. Thats how I dont get the asnwer to my question from the PDF.

What I expect:
I expect to get the answer of my question from the uploaded PDF.

What I have selected as parameters for RAG:
In Admin Panel -> Settings -> Documents:
• Content Extraction - Default
• PDF Extract Images (OCR) : Default
• Bypass Embedding and Retrueval : Off
• Text Splitter: Default (Character)
• Chunk Size: 1000 with 100 as a overlap
• Embedding Model Engine: Default (Setence Transformer) sentence-transformers/all-MiniLM-L6-v2
• Full Context Mode: Off
• Hybrid Search: Tried both On / Off, but when on, i used below params
- Reranking model: BAAI/bge-reranker-v2-m3
- Top K: 5
- Top K reranker: 5
- Relevance Threshold: 0
• RAG Template:
### Task:
Respond to the user query using the provided context, incorporating inline citations in the format [source_id] only when the <source_id> tag is explicitly provided in the context.

	### Guidelines:
	- If you don't know the answer, clearly state that.
	- If uncertain, ask the user for clarification.
	- Respond in the same language as the user's query.
	- If the context is unreadable or of poor quality, inform the user and provide the best possible answer.
	- If the answer isn't present in the context but you possess the knowledge, explain this to the user and provide the answer using your own understanding.
	- **Only include inline citations using [source_id] (e.g., [1], [2]) when a `<source_id>` tag is explicitly provided in the context.**
	- Do not cite if the <source_id> tag is not provided in the context.  
	- Do not use XML tags in your response.
	- Ensure citations are concise and directly related to the information provided.
	
	### Example of Citation:
	If the user asks about a specific topic and the information is found in "whitepaper.pdf" with a provided <source_id>, the response should include the citation like so:  
	* "According to the study, the proposed method increases efficiency by 20% [whitepaper.pdf]."
	If no <source_id> is present, the response should omit the citation.
	
	### Output:
	Provide a clear and direct response to the user's query, including inline citations in the format [source_id] only when the <source_id> tag is present in the context.
	
	<context>
	{{context}}
	</context>
	
	<user_query>
	{{question}}
	</user_query>
• LLM model selected in Open WebUI: qwen2:72b 
• Open WebUI version: 0.6.7
• Ollama Version:  0.6.5

Please give me some insights how I can enable a successfull RAG that communicate with the document and retrieve a good answers.

Actual Behavior

It behaves like,
When I upload the document and ask the question from the PDF, it does not even give me a correct answer. I am afriad does the chatbot even read the data from the document, chunk it and make it into the vector storage or not.

Same happens when I create the Knowledge Base and then I create the Model (Considering the Knowledge Base) and asks the Document about my question.

Steps to Reproduce

Click on Workspace - > Knowlegde -> Create Knowledge -> Upload the File

Click on Workspace - > Model-> Name the Model -> write the prompt -> define advance parameters -> add Knowledge Base to that Model.

Go to your main Screen of Chatbot, write '#' and select the Knowledge Base and Ask the question.

Logs & Screenshots

Additional Information

No response

Originally created by @Urvesh71 on GitHub (May 12, 2025). Original GitHub issue: https://github.com/open-webui/open-webui/issues/13807 ### Check Existing Issues - [x] I have searched the existing issues and discussions. - [x] I am using the latest version of Open WebUI. ### Installation Method Docker ### Open WebUI Version 0.6.7 ### Ollama Version (if applicable) 0.6.5 ### Operating System Ubuntu ### Browser (if applicable) _No response_ ### Confirmation - [x] I have read and followed all instructions in `README.md`. - [x] I am using the latest version of **both** Open WebUI and Ollama. - [x] I have included the browser console logs. - [x] I have included the Docker container logs. - [x] I have listed steps to reproduce the bug in detail. ### Expected Behavior While I do perform RAG with Document in Open WebUI, I am not getting a correct answers. I do not know what is wrong with it. When I upload the PDF and ask the question, everyatime it answers that there is not content provided. Thats how I dont get the asnwer to my question from the PDF. What I expect: I expect to get the answer of my question from the uploaded PDF. What I have selected as parameters for RAG: In Admin Panel -> Settings -> Documents: • Content Extraction - Default • PDF Extract Images (OCR) : Default • Bypass Embedding and Retrueval : Off • Text Splitter: Default (Character) • Chunk Size: 1000 with 100 as a overlap • Embedding Model Engine: Default (Setence Transformer) sentence-transformers/all-MiniLM-L6-v2 • Full Context Mode: Off • Hybrid Search: Tried both On / Off, but when on, i used below params - Reranking model: BAAI/bge-reranker-v2-m3 - Top K: 5 - Top K reranker: 5 - Relevance Threshold: 0 • RAG Template: ### Task: Respond to the user query using the provided context, incorporating inline citations in the format [source_id] **only when the <source_id> tag is explicitly provided** in the context. ### Guidelines: - If you don't know the answer, clearly state that. - If uncertain, ask the user for clarification. - Respond in the same language as the user's query. - If the context is unreadable or of poor quality, inform the user and provide the best possible answer. - If the answer isn't present in the context but you possess the knowledge, explain this to the user and provide the answer using your own understanding. - **Only include inline citations using [source_id] (e.g., [1], [2]) when a `<source_id>` tag is explicitly provided in the context.** - Do not cite if the <source_id> tag is not provided in the context. - Do not use XML tags in your response. - Ensure citations are concise and directly related to the information provided. ### Example of Citation: If the user asks about a specific topic and the information is found in "whitepaper.pdf" with a provided <source_id>, the response should include the citation like so: * "According to the study, the proposed method increases efficiency by 20% [whitepaper.pdf]." If no <source_id> is present, the response should omit the citation. ### Output: Provide a clear and direct response to the user's query, including inline citations in the format [source_id] only when the <source_id> tag is present in the context. <context> {{context}} </context> <user_query> {{question}} </user_query> • LLM model selected in Open WebUI: qwen2:72b • Open WebUI version: 0.6.7 • Ollama Version: 0.6.5 Please give me some insights how I can enable a successfull RAG that communicate with the document and retrieve a good answers. ### Actual Behavior It behaves like, When I upload the document and ask the question from the PDF, it does not even give me a correct answer. I am afriad does the chatbot even read the data from the document, chunk it and make it into the vector storage or not. Same happens when I create the Knowledge Base and then I create the Model (Considering the Knowledge Base) and asks the Document about my question. ### Steps to Reproduce Click on Workspace - > Knowlegde -> Create Knowledge -> Upload the File Click on Workspace - > Model-> Name the Model -> write the prompt -> define advance parameters -> add Knowledge Base to that Model. Go to your main Screen of Chatbot, write '#' and select the Knowledge Base and Ask the question. ### Logs & Screenshots ![Image](https://github.com/user-attachments/assets/28ef85e4-a6df-4ce8-a96a-da0ba718d53d) ### Additional Information _No response_

GiteaMirror added the bug label 2026-04-25 06:30:06 -05:00

GiteaMirror closed this issue

2026-04-25 06:30:07 -05:00

Sign in to join this conversation.

Branches Tags

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/open-webui#32569