Open-Source Visual AI Workflow Builder

Build AI Workflows Visually

Connect nodes to generate content, analyze data, and automate tasks. Run models locally or via cloud APIs.

Start Here

NodeTool is a visual workflow builder for AI pipelines—connect nodes for images, video, text, data, and automation. Run locally or deploy to RunPod, Cloud Run, or self-hosted servers.

Who uses NodeTool?

Creators & Designers

Generate and transform media with AI.

Use Flux, Qwen Image, and custom models
Generate variations
Build reusable workflows

Developers & Researchers

Build agents, RAG systems, and pipelines.

Design multi-step LLM agents
Index and query documents locally
Deploy workflows as APIs

Data & Business Users

Process documents and automate tasks without coding.

Process data with AI pipelines
Automate document workflows
Build custom tools

What you can build right away

AI Agents & Automation

Build multi-step agents that plan and execute tasks.

Agent patterns → Realtime Agent →

Document Intelligence & RAG

Index documents, search with AI, and answer questions from sources.

RAG pattern → Chat with Docs →

Image & Video Creation

Generate and transform media with AI models.

Movie Posters → Story to Video →

Data Processing & Analysis

Transform data, extract information, and automate reports.

Data Visualization → Data patterns →

Your first 10 minutes

Download NodeTool — install for macOS, Windows, or Linux.
Choose your AI models — install local models like Flux or Qwen Image, or use cloud services.
Try a template workflow — explore examples, press Run, watch results stream.
Experiment and customize — change inputs, connect nodes differently, make it yours.
Share your workflow — turn it into a simple app others can use.

Choose your path

I’m new to NodeTool: Getting Started, Workflow Editor, Image Editor, Hardware Requirements, Example Gallery.
I want to explore: Example Workflows, Workflow Patterns, Community.
I need AI models: Models & Providers, HuggingFace Integration, Local vs Cloud.
I’m coming from another tool: How NodeTool Compares – understand where NodeTool fits.
I’m ready to deploy: Deployment Guide, Self-Hosted Setup, API Access, Configuration.
I’m a developer: Developer Guide, Custom Nodes, CLI Reference, API Reference.
Reference & Help: Workflow Debugging, Troubleshooting, Glossary, Node Library.

Local-first or cloud-augmented

Local-only mode

All workflows, assets, and models run on your machine for maximum privacy and control.

Use MLX, llama.cpp, Whisper, and Flux locally
Store assets on disk or your own storage
Work offline once models are downloaded

Storage options →

Cloud-augmented mode

Mix local AI with cloud services for flexibility. Use the best tool for each task.

Add API keys for OpenAI, Anthropic, Replicate
Access the latest cloud models on demand
Deploy workflows to cloud infrastructure

Models & Providers →

Key Features

🎯 Visual Editor

Build workflows by connecting nodes. No coding required. View the entire pipeline on one canvas.

👁️ Real-time Debugging

See every step execute in real-time. Inspect intermediate outputs. Streaming execution shows progress as it happens.

🔒 Local Execution

Run LLMs, Whisper, and diffusion models on your infrastructure. Cloud APIs are opt-in.

Popular destinations

Build with the community

NodeTool is open-source under AGPL-3.0. Join the Discord, explore the GitHub repo, and share workflows with other builders.