FlowForge - Portfolio
FlowForge
AI Systems

FlowForge

A RAGFlow fork with a custom interface and changes to document ingestion and retrieval across text, tables, and images.

Extends an upstream project with a window-based interface and changes to ingestion and retrieval.

Overview

FlowForge extends RAGFlow with a custom interface and changes to document ingestion and retrieval across text, tables, and images.

What’s different

  • Custom front-end built around a window-management UX rather than a flat dashboard.
  • Native support for vision-language models (VLMs) in the retrieval path — image chunks are first-class citizens alongside text.
  • Re-engineered ingestion pipeline with pluggable chunkers and changes to table extraction.
  • Continued API compatibility with upstream RAGFlow where it makes sense.

Architecture highlights

The core stays Python (FastAPI) for the API layer and worker pool, with ElasticSearch for hybrid retrieval, MinIO for object storage, and Redis for the queue. The front-end is a TypeScript / React SPA that talks to a stable REST API. The VLM inference is dispatched to a configurable backend (open-weights served locally, or a managed provider via a thin adapter).

Limitations

Source is not publicly available. The on-page demo uses illustrative retrieval scenarios.

Scripted demonstration. This walkthrough uses example data and does not call live services or run operations on your systems.

Illustrative scenarios show retrieved text, tables, and image evidence in the fork's workflow.

Initializing demo…