OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
RAG & Knowledge Systems

Pdfllm

Pdfllm is an open-source Python RAG microservice that processes and indexes documents for semantic and knowledge-graph search, then generates answers with cited sources. It combines PostgreSQL session and metadata storage, Qdrant vector storage, Dgraph graph indexing and Celery asynchronous processing.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Pdfllm — OpenViST project profile in RAG & Knowledge Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you can self-host a Python service and need document ingestion, hybrid retrieval and LLM-based Q&A over a managed collection.

Who it is for

  • Teams or developers who want to operate a document-search and question-answering backend.
  • Organizations that need to categorize business documents and query a category or a broader collection.

Documented uses

  • Parse and index documents, including scanned images and PDFs, for retrieval.
  • Search a collection using semantic and graph-based retrieval.
  • Ask questions about a document set or a selected category and receive answers with cited sources.
  • Preview uploaded files and organize material with custom categories and prompts.
Before you choose

Requirements & limits

The repository is public and identifies the project as open source under the MIT license. The documented deployment requires an OpenAI API key.

  • The documented stack includes PostgreSQL, Qdrant, Dgraph and Celery, in addition to the application itself.
  • The README lists support for PDF, TXT, DOC/DOCX, XLS/XLSX, CSV, common image formats, Markdown, RTF and OpenDocument formats.
  • Requests to documented endpoints require X-API-Key authentication.
  • The README says PDFs over 40 pages may face context-window limitations; multi-file handling is reported as tested up to five files.
  • The Master Chat orchestration chain is described as in progress, and the README says a RAG evaluation framework is not yet implemented.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Deploy the service with Docker Compose

Configure and launch the documented pdfLLM application stack.

Before you start

  • A local clone of the pdfLLM repository.
  • Docker Compose.
  • An OpenAI API key and PostgreSQL settings for the environment file.

Steps

  1. Clone the repository and change into the pdfLLM directory.

  2. Copy env_example to .env.

  3. Set the OpenAI API key, PostgreSQL settings and other required configuration in .env.

  4. Run docker compose up -d --build.

Expected outcome

The Docker Compose application is launched using the configured environment.

Guide limitations

  • This setup procedure does not document a subsequent request or ingestion workflow.
  • The README notes that PDFs over 40 pages may encounter context-window limitations and multi-file processing is reported as tested up to five files.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
ikantkode/pdfLLM · Official homepage title · retained in its source language
How we review sourcesSuggest a correction