LibreAI / Automation / ai-flow
// automation Β· DahnM20/ai-flow

ai-flow

Drag-and-drop builder for custom AI workflows across your favorite models.

πŸ’Ž Hidden gemβ˜… 291 starsβ‘‚ 54 forksMITupdated 2026-05
β–¦ visual β€” this project's README ships no embeddable screenshot,
so here's a placeholder. The guide below is complete regardless.
// project facts

Project facts

ai-flow is an open-source automation project by DahnM20 (DahnM20/ai-flow on GitHub). It has 291 GitHub stars and 54 forks, is released under the MIT license, and was last updated on GitHub on 26 May 2026.

Repository
DahnM20/ai-flow
GitHub stars
291
Forks
54
License
MIT
Last updated on GitHub
26 May 2026
Category
Automation
Listing
Hidden gem β€” a shorter page for a lesser-known project
Official site
Official site
// how it works

How it works

  1. Open the canvas

    Launch AI-Flow and start with a blank workflow.

  2. Drop in model nodes

    Add text, image, or chat models from the node palette.

  3. Connect the flow

    Wire nodes together so outputs feed the next step.

  4. Run with your keys

    Add your own API keys and execute the workflow.

// features

Why people love it

Drag-and-drop canvas

Build workflows visually, no code required.

Multi-model support

Chain GPT-4o, Stable Diffusion, FLUX, and more.

Bring your own keys

Connect your own API accounts to each model.

Node-based logic

Branch, merge, and sequence AI steps freely.

Self-hostable

Run it yourself and keep data in your hands.

MIT licensed

Fully open source under the MIT license.

Advertisement
// run it

Run it yourself

  1. Step 1 β€” Clone the repo

    Download the source and move into the project folder.

    git clone https://github.com/DahnM20/ai-flow.git
    cd ai-flow
  2. Step 2 β€” Install dependencies

    Follow the README install step for the Node stack.

    npm install
  3. Step 3 β€” Launch it

    Start the app and open the workflow canvas.

    npm run dev
// use cases

What you can build

Advertisement
// faq

Questions, answered

Is AI-Flow free?

Yes β€” MIT licensed and free to run; you pay only for the model APIs you connect.

Which models can I use?

GPT-4o, Stable Diffusion, FLUX, and more via your own API keys.

Do I need to code?

No β€” the drag-and-drop canvas is built for non-coders.

Can I self-host it?

Yes, it is designed to be self-hosted.

Where do my API keys live?

In your own instance β€” nothing is shared with third parties.

// keep exploring

Similar tools

Automation

n8n

Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+…

β˜… 206.6k starsCustom (check repo)
View guide β†’
Automation

automa

A browser extension for automating your browser by connecting blocks

β˜… 21.6k starsUnknown
View guide β†’
Automation

Windmill

Open-source developer platform to power your entire infra and turn scripts into webhooks, workflows and UIs. Fastest workflow engine…

β˜… 18.1k starsCustom (check repo)
View guide β†’
Automation

Langflow

Langflow β€” the visual builder for AI agents and workflows: drag-and-drop flows over LangChain components, deployed as APIs or MCP tools, MIT-licensed

β˜… 155.8k starsMIT
View guide β†’
Automation

workflowbuilder

React SDK for embedding drag-and-drop workflow editors in your product.

β˜… 388 starsApache-2.0
View guide β†’
Automation

flow-like

Strongly typed, enterprise-scale workflows built for speed and AI integration.

β˜… 957 starsCustom
View guide β†’
// alternatives

Alternatives in Automation

The most-starred other Automation projects in the LibreAI catalogue β€” the realistic shortlist if ai-flow is not the right fit.

Get ai-flow

Free and open β€” grab the source, star the repo, and start creating.