Text-to-Image Node

Generate images from text descriptions using 30+ AI models.

Overview

The Text-to-Image node creates images from text prompts. It's one of the most versatile nodes in Canvas, supporting multiple AI models with different styles and capabilities.

Inputs

Text Prompt (required)

Type or connect from Prompt Node
Describes what you want to create
More detail = better results

Outputs

Generated Images

1-4 images per generation
Saved to Media Library automatically
Can connect to other nodes

Configuration

Model Selection

Choose from 30+ models:

FLUX Models:

FLUX Schnell - Fast, good quality
FLUX Dev - Higher quality, slower
FLUX Pro - Professional quality

Leonardo Models:

Phoenix - Versatile, high quality
Kino - Cinematic style
Various LoRA models - Specialized styles

Other Models:

Stable Diffusion variants
Specialized art models
Style-specific models

Aspect Ratio

Choose image dimensions:

Square (1:1)
Portrait (3:4, 9:16)
Landscape (16:9, 21:9)
Custom

Quality Settings

Number of Images: 1-4 images per generation

Quality Level: (model-dependent)

Draft
Standard
High Quality

Advanced Settings

Seed - Reproducible results
Guidance - How closely to follow prompt
Steps - Quality vs speed trade-off

Usage Examples

Basic Image Generation

Prompt: "A serene mountain landscape at sunset"
Model: FLUX Schnell
Aspect: 16:9
Result: Landscape photo

Character Design

Prompt: "Futuristic robot character, detailed mechanical parts"
Model: FLUX Dev
Aspect: 3:4
Result: Character concept art

Product Photography

Prompt: "Modern smartwatch on marble surface, studio lighting"
Model: Leonardo Phoenix
Aspect: 1:1
Result: Product photo

Workflow Integration

Common Connections

Input from:

Prompt Node (text prompts)
Chat Node (AI-refined prompts)

Output to:

Image-to-Video (animate the image)
Image-to-Image (create variations)
Multi-Angle Generator (different views)
Image-to-3D (3D model)

Tips & Best Practices

Write Better Prompts:

Be specific about style and details
Include lighting, mood, perspective
Mention art style or artist references

Model Selection:

Test: FLUX Schnell
Production: FLUX Dev or Leonardo Phoenix
Style-specific: Use appropriate LoRA

Optimize Speed:

Use Schnell for drafts
Generate 1 image first, then more if needed
Lower quality settings for testing

Troubleshooting

Poor quality results:

Try more detailed prompt
Switch to higher quality model
Increase quality settings

Generation failed:

Check prompt for banned content
Try different model
Simplify complex prompts

Wrong style:

Add style keywords to prompt
Try different model
Use Leonardo for specific styles

Cost & Usage

Counts as 1 image per generated image
4 images = 4 units used
Different models may affect overage costs

Learn about pricing →

Related Nodes

Image-to-Image - Modify generated images
Image-to-Video - Animate images
Prompt Node - Store reusable prompts