yuqie6/ProductFlow

gpt-image-2 画图工作台 / Self-hosted workbench for AI copy, posters, image sessions, and visual product workflows

What it solves

ProductFlow provides a self-hosted workspace for solo entrepreneurs and small merchant teams to manage product assets. It streamlines the process of creating product materials by integrating product data, reference images, AI-generated copy, and AI-generated posters into a single visual workflow, eliminating the need to jump between disparate AI tools for e-commerce content creation.

How it works

The system uses a node-based canvas where users organize product information and connect different AI tasks. It integrates with text and image model providers (such as OpenAI) to automate the generation of marketing copy and visuals. The backend is powered by FastAPI and Dramatiq for asynchronous task processing, while the frontend is a React-based interface that supports both desktop and mobile interactions for managing the canvas and image generation sessions.

Who it’s for

It is designed for individual merchants or small e-commerce teams who need a private, self-hosted solution to generate and organize product marketing assets at scale.

Highlights

  • Visual Workflow Canvas: A node-based editor to organize product data, reference images, and AI generation steps with support for templates.
  • Integrated AI Copy & Image Gen: Dedicated nodes for generating structured marketing text and images, including support for continuous image generation sessions.
  • Flexible Poster Generation: Supports both template-based rendering (via Pillow) and AI-generated posters.
  • Self-Hosted Infrastructure: Deployable via Docker Compose with a full stack including PostgreSQL, Redis, and a dedicated worker for background tasks.
  • Comprehensive Asset Gallery: A centralized gallery to save, track, and manage all generated results across different sessions and products.

Related

  • Project
    hassancs91/claude-youtube-editor

    An open-source pipeline that automates YouTube video editing by turning raw talking-head recordings into polished videos with AI-generated visuals, audio cleaning, and automated uploads.

  • Project
    pwilkin/trellis.cpp

    A standalone C++ implementation of the TRELLIS image-to-3D pipeline using GGML, enabling the generation of textured 3D models from images without Python at runtime.

  • Project
    huggingface/huggingface-gemma-recipes

    A collection of minimal recipes and notebooks for implementing multimodal inference, fine-tuning, and RAG with the Gemma family of models.

  • Project
    horang-labs/tessera

    Tessera is a local desktop/web app that lets you run multiple Claude Code, Codex, or OpenCode AI coding agents in parallel, each in its own isolated Git worktree. It provides a Kanban board, split‑pane UI, chat view for terminals, full Git integration, and mobile remote access, all while using the provider CLIs already installed on your machine.