Top 200 ReposEXPLORER
Browse agents, categories and pages
Back to results
Reported by sourceThe source says this happened. Top 200 Repos has not reproduced it.

Build a cited Q&A knowledge base over your own documents

Self-host RAGFlow, upload PDFs, Office files and scans, and ask questions that come back with traceable source chunks.

Try the promptRead the original (opens in a new tab)

The problem

How it works

You start RAGFlow with Docker Compose, then add a chat model, an embedding model and a reranker in the web UI. Documents such as Word, PowerPoint, Excel, PDFs, scans and web pages are parsed with deep document understanding and template-based chunking. Answers include references to the source chunks, and agentic retrieval offers several thinking modes for harder questions.

Results and limitations

The maintainers claim the visible chunks and traceable citations make answers easier to check and reduce hallucinations.

Before you try it

The suggested starting point is 4 CPU cores, 16 GB RAM and 50 GB disk, with Docker 24+; Elasticsearch needs a kernel setting change. Model providers you configure may charge per call. Results depend on your setup and were not independently reproduced by Top 200 Repos.

Tools and repositories

Tools
Docker 24+ · Docker Compose v2.26+
Prerequisites
A host with roughly 16 GB RAM and API keys or local endpoints for chat, embedding and reranker models.

Try it yourself: Plan a RAGFlow knowledge base

Fill the fields to adapt the prompt. Empty fields stay as visible [placeholders].

Preview
Help me set up a RAGFlow knowledge base for [Use case] using the documents in [Document location]. First check this host meets the documented CPU, RAM, disk and Docker requirements and tell me what is missing. Recommend a chunking template for each document type ([Document types]) and explain why. Propose a chat, embedding and reranker model, using [Model preference], and wait for my approval before entering any keys or uploading files. After ingestion, write ten test questions with expected source documents, run them, and report which answers cited the right chunks.

Keep sensitive documents on a private instance and spot-check citations before trusting answers.

Related videos

About infiniflow/ragflow

Independent creators. Listing a video implies no endorsement, sponsorship or partnership.

Sources

  1. infiniflow/ragflow README (opens in a new tab)

    Project documentation · github.com · Undated · Source checked 9 Nov 2026, 22:23 UTC

    What this source supports
    • Deep document understanding and template-based chunking
    • Grounded citations with viewable source chunks
    • Agentic retrieval with selectable thinking modes
    • Docker Compose start; 4 cores, 16 GB RAM, 50 GB disk suggested
    • Configure chat, embedding and reranker models in the UI
Report a problem

Reported by source

Answer technical questions by letting async agents run experiments

Simon Willison sends research questions to cloud coding agents such as Claude Code for web and Codex Cloud, which write and run experiments in a dedicated repo and open a PR.

simonwillison.net · 6 Nov 2025anthropics/claude-code, openai/codex
Reported by source

Let a coding agent run overnight ML training experiments

Point a coding agent at autoresearch so it edits a small GPT training script, runs short timed experiments and keeps what improves.

github.comkarpathy/autoresearch, anthropics/claude-code, openai/codex
Planned, not done

Run long research tasks with sub-agents in DeerFlow

DeerFlow 2.0 is an open-source agent harness from ByteDance that plans a task, hands parts to sub-agents and works in a sandbox with memory, skills and search tools, using any OpenAI-compatible model.

github.combytedance/deer-flow
Reported by source

Run unattended ML training experiments with autoresearch

Red Hat's Roy Belio containerized Karpathy's autoresearch on OpenShift AI with H100 GPUs and let Claude Code run 198 training experiments over 24 hours.

developers.redhat.com · 7 Apr 2026karpathy/autoresearch, anthropics/claude-code
Planned, not done

Build a support chatbot that answers from your help docs with Dify

Turn help-centre pages into a Dify knowledge base, route questions with a classifier, answer from retrieved context and embed the bot on your website.

docs.dify.ailanggenius/dify, firecrawl/firecrawl
Reported by source

Crawl a docs site into clean Markdown for a RAG index

Use Firecrawl to map and crawl a website, returning each page as Markdown or structured JSON ready to chunk and embed.

github.comfirecrawl/firecrawl