Skip to content

itsevin/ImageFlow-Auto

Repository files navigation

ImageFlow Auto

ImageFlow Auto logo

Recoverable batch image workflow console.
Turn prompt batches, provider tasks, review records, cost estimates, and export packages into one inspectable run.

English · 简体中文

License No API local run Python React Recoverable runs

ImageFlow Auto console

Choose Your README

Language File What you get
English README.en.md Full overview, quick start, screenshots, architecture, pricing, provider model, FAQ, roadmap.
简体中文 README.zh-CN.md 完整中文介绍、快速启动、截图、架构、计费、Provider、FAQ、路线图。

Documentation

Fastest Local Run

python -m pip install -e .
cd web
npm ci
npm run build
cd ..
$env:IMAGEFLOW_RUNTIME_ROOT="C:\temp\imageflow-demo"
python -m imageflow_auto run --track high_res --works 1 --candidates 4 --finals 4 --providers mock_image --mock-design --mock-vlm
python -m imageflow_auto console --host 127.0.0.1 --port 8765

Open http://127.0.0.1:8765.

Single-brief and ultra-resolution runs are also supported:

python -m imageflow_auto run --track ultra_res --resolution wide_8k --title "Future museum systems" --requirements "Museum operations studio with archive tables and visitor flow." --candidates 4 --finals 4 --providers mock_image --mock-design --mock-vlm

What Is This?

ImageFlow Auto is a public, reusable workflow runner for AI image batches: prompt batches, provider adapters, generated files, review state, pricing estimates, and export packages are handled as one recoverable pipeline.

Runs can start from safe sample topics, a single typed brief, or batch-imported briefs. The ultra_res track supports named resolution presets and custom width/height with whole-image enhancement before quality checks.

It is not a hosted SaaS, not a payment platform, and not a private campaign dump. The public project is designed to be run safely with mock_image before you connect any real provider.

Provider configuration is role-based: image generation, design planning, quick review, structure review, group review, and final review can each use their own ordered fallback chain. The web console writes local overrides to ignored files, so public source stays publishable while real local runs can use hosted APIs.

About

A local-first AI batch image workflow console for generation, review, cost tracking, recovery, and export.

Resources

License

Contributing

Security policy

Stars

1 star

Watchers

0 watching

Forks

Packages

 
 
 

Contributors