Skip to main content

Flow Editor

The Flow Editor is DUTO's visual workflow builder. Create powerful content generation pipelines by connecting nodes in a visual graph.

What is the Flow Editor?​

The Flow Editor lets you build automated workflows without writing code. Connect specialized nodes to:

  • Generate images and videos using AI
  • Process and transform media
  • Add intelligence and decision-making
  • Batch process large amounts of content
  • Save and organize outputs

Key Concepts​

Nodes​

Nodes are the building blocks of your workflow. Each node performs a specific function:

  • Input nodes - Start your workflow or provide data
  • Processing nodes - Transform, generate, or analyze content
  • Output nodes - Save or display results

DUTO includes 50 specialized nodes across categories like text, image, video, audio, and intelligence.

Browse all nodes →

Edges (Connections)​

Edges connect nodes together, defining how data flows through your workflow. Data types are color-coded:

  • image Blue - Static images
  • video Purple - Video files
  • audio Green - Audio files
  • text Yellow - Text/prompts
  • json Orange - Structured data
  • any Gray - Universal connector

Workflows​

A workflow is a complete graph of connected nodes that accomplishes a task. Workflows can be:

  • Saved and reused from your projects dashboard
  • Run on demand with visual feedback
  • Parameterized for different inputs using data feeds

Workflow Execution​

When you run a workflow, DUTO executes nodes in dependency order:

  1. Nodes wait for all inputs to be ready
  2. Independent branches execute in parallel
  3. Progress is shown visually in real-time
  4. Results flow through connections to the next nodes

For example, a workflow might fan out from a starting node into two parallel branches (Node A and Node C), then merge into a final node (Node B) that waits for both branches to finish before running.

In this example, Nodes A and C can run in parallel, and Node B waits for both to complete.

Features​

Visual Building​

  • Drag-and-drop node placement
  • Click-to-connect edges
  • Pan and zoom the canvas
  • Mini-map navigation
  • Quick slots for frequently used nodes

AI-Assisted Creation​

Let AI help build your workflows:

  • Describe what you want to create
  • AI generates a complete workflow
  • Refine and customize as needed

Learn about AI generation →

Real-Time Feedback​

Watch your workflow execute:

  • Node status indicators (waiting, running, complete, error)
  • Animated edge flow
  • Live previews of generated content
  • Execution timing and credit usage

Batch Processing​

Process multiple items through the same workflow:

  • BatchProcessor and BatchEnd nodes
  • Parallel or sequential processing
  • Progress tracking and statistics
  • Error handling options

Getting Started​

  1. Interface Overview - Learn the UI
  2. Creating Workflows - Build your first workflow
  3. Executing Workflows - Run and debug
  4. Node Reference - Explore all 50 nodes

Example Workflows​

Simple Image Generation​

  1. Start — Begin the workflow
  2. Text Prompt — Type your description, e.g. "A golden sunset over the ocean, cinematic"
  3. Text to Image — AI generates the image using your chosen model (SeedDream, Flux Kontext, etc.)
  4. Preview — View, download, or send to the Content Library

Image Variations with Batch​

  1. Start — Begin the workflow
  2. Data Feed — Load a list of prompts or items to process
  3. Batch Processor — Loops through each item, running the chain once per entry
  4. Text to Image — Generates an image for each item in the feed
  5. Batch End — Collects all results with timing stats
  6. Preview — Review all generated images at once

Video with Audio​

  1. Start — Begin the workflow
  2. Text Prompt — Describe your scene, e.g. "A timelapse of storm clouds rolling over mountains"
  3. Text to Video (Veo 3) — Generates a video clip with native audio using Google's Veo 3 model
  4. Preview — Watch and download the video

Intelligent Content​

  1. Start — Begin the workflow
  2. Upload Media — Provide a reference image or video
  3. Brain (Analysis mode) — Analyzes the media and extracts a detailed description
  4. Text to Image — Uses the extracted description to generate a new image
  5. Preview — Compare the reference and generated result side by side