Skip to main content

Your First Workflow

Learn the basics of the Flow Editor by creating a simple image generation workflow.

What We'll Build​

In this tutorial, you'll create a workflow that:

  1. Takes a text prompt
  2. Generates an image using AI
  3. Saves the result to your library

Step 1: Create a New Workflow​

  1. From your dashboard, click New Workflow
  2. Give your workflow a name (e.g., "My First Workflow")
  3. Click Create

You'll now see the Flow Editor canvas with a Start node already placed.

Step 2: Add a Text Prompt Node​

The Text Prompt node lets you define what image to generate.

  1. Open the node menu at the bottom of the canvas
  2. Search for "Text Prompt" or browse the Text category
  3. Click to add the node to your canvas
  4. Connect the Start node's output to the Text Prompt node's input

Configure the Node​

  1. Click on the Text Prompt node to select it
  2. In the node panel, enter your prompt: 
    A serene mountain landscape at sunset, with a calm lake
    reflecting the orange and purple sky

Step 3: Add an Image Generation Node​

Now let's generate an image from your prompt.

  1. Open the node menu and search for "Text to Image"
  2. Add the Text to Image node to your canvas
  3. Connect the Text Prompt's output to the Text to Image input

Configure the Node​

  1. Select the Text to Image node
  2. Choose your preferred model (e.g., "Nano Banana Pro" or "SeedDream")
  3. Set the aspect ratio (e.g., "16:9" for landscape)
  4. Leave other settings as default for now

Step 4: Add a Preview Node​

To see your generated image:

  1. Add a Preview Media node from Media I/O → Preview Media
  2. Connect the Text to Image output to the Preview Media input

Step 5: Run Your Workflow​

  1. Click the Run button in the toolbar (or press Cmd/Ctrl + Enter)
  2. Watch as each node executes:
    • The Start node turns green
    • The Text Prompt processes your text
    • The Text to Image node generates your image (this takes a few seconds)
    • The Preview displays your result

Understanding the Visual Feedback​

As your workflow runs, you'll see:

  • Gray - Node is waiting
  • Blue - Node is currently executing
  • Green - Node completed successfully
  • Red - Node encountered an error

The edges (connections) also animate to show data flowing through your workflow.

Your Complete Workflow​

Your finished workflow should look like this:

Start → Text Prompt → Text to Image → Preview Media

Generated images are automatically saved as artifacts in your workflow and can be accessed from the Content Library.

Tips for Success​

Experiment with prompts

Try different prompts to see how the AI interprets your descriptions. Be specific about:

  • Subject matter
  • Style (photorealistic, illustration, etc.)
  • Lighting and mood
  • Color palette
Try different models

Each image model has different strengths:

  • Nano Banana Pro - Fast, high quality generation by Google
  • SeedDream - Budget-friendly option by ByteDance
  • Flux Kontext - Versatile with LoRA support

What's Next?​

Now that you've created your first workflow, explore: