DEDUPED Hot AI repos, turned into use cases Get the Monday 5

How to chat with your own documents using local RAG with Open WebUI

Load documents into a chat or your library and ask questions answered from them.

Short answer: Upload documents to the library or a chat; then Pull them into a conversation with the # command; ask questions; retrieval uses hybrid BM25 and vector search with reranking. All steps.

Who it is forA researcher or small team that wants answers grounded in their own files without a cloud service.

The problemCloud document chat tools require uploading private files.

The outcomeAnswers drawn from your documents with hybrid search and reranking, on your server.

What you need first

  • Docker, or Python with pip
  • Ollama for local models, or an OpenAI-compatible API key
  • A GPU is optional (the :cuda image uses it)

Steps

  1. Upload documents to the library or a chat
  2. Pull them into a conversation with the # command
  3. Ask questions; retrieval uses hybrid BM25 and vector search with reranking
  4. Add web search from a provider such as SearXNG if you want fresh results too

Prompt for your agent

In my open-webui/open-webui, create a knowledge collection from <documents>, then answer <question> citing which document each point came from.

Caveat.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for open-webui/open-webui and "Chat with your own documents using local RAG"

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Share: X LinkedIn Hacker News Email Copy link

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer