Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.opencanva.co/llms.txt

Use this file to discover all available pages before exploring further.

​
Introduction to prompts

A prompt is the text description you provide to tell the AI what image to generate. The quality and detail of your prompt directly impacts the quality of your generated image.

​
Prompt structure

Effective prompts typically include these elements:
  1. Subject: What is the main focus of the image?
  2. Style: What artistic style or aesthetic?
  3. Composition: How is the image framed?
  4. Details: Colors, lighting, mood, background
  5. Quality modifiers: Terms that enhance output quality

​
Writing techniques

​
Be specific and descriptive

Weak prompt: 
A cat
Strong prompt: 
A fluffy orange tabby cat sitting on a windowsill, golden hour lighting, soft focus, cozy home interior, warm color palette

​
Include style and mood

Add style descriptors to guide the aesthetic:
  • Artistic styles: “watercolor painting”, “digital art”, “photorealistic”, “sketch”
  • Mood: “serene”, “dramatic”, “playful”, “mysterious”
  • Era/period: “medieval”, “futuristic”, “vintage 1950s”

​
Describe composition

Tell the AI how to frame the image:
  • Shot type: “close-up”, “wide shot”, “portrait”, “landscape view”
  • Angle: “bird’s eye view”, “low angle”, “eye level”
  • Focus: “shallow depth of field”, “everything in focus”

​
Add technical details

Include lighting, colors, and technical aspects:
  • Lighting: “soft natural light”, “dramatic shadows”, “neon lighting”
  • Colors: “warm color palette”, “monochrome”, “vibrant colors”
  • Quality: “high detail”, “sharp focus”, “professional photography”

​
Example prompts

​
Landscape photography

A serene mountain landscape at sunset, snow-capped peaks in the background, a calm alpine lake in the foreground reflecting the orange and pink sky, dramatic clouds, golden hour lighting, photorealistic, high detail, 4K quality

​
Character design

A friendly robot character, round and approachable design, blue and white color scheme, large expressive eyes, standing pose, clean vector art style, modern illustration, bright and cheerful mood

​
Abstract art

Abstract geometric patterns, vibrant gradient colors transitioning from purple to pink to orange, flowing organic shapes, digital art style, high contrast, dynamic composition, modern aesthetic

​
Product visualization

A minimalist coffee mug on a wooden table, soft natural lighting from the left, shallow depth of field, white background, product photography style, clean and professional, high resolution

​
Advanced techniques

​
Using style references

Reference specific artistic styles: 
A cyberpunk cityscape, inspired by Blade Runner aesthetic, neon signs in Japanese and English, rain-soaked streets, flying vehicles, night scene, cinematic lighting, highly detailed

​
Combining concepts

Merge different ideas creatively: 
A steampunk version of a Victorian library, brass gears and clockwork mechanisms integrated into bookshelves, warm gas lamp lighting, cozy atmosphere, detailed illustration style

​
Negative prompts

While Open Canva doesn’t use explicit negative prompts, you can guide away from unwanted elements by being specific about what you want: Instead of: “A cat, not a dog” Better: “A cat sitting on a windowsill” (specificity naturally excludes dogs)

​
Common mistakes to avoid

Too vague: “Something cool” won’t produce good results. Be specific.
Contradictory terms: Avoid mixing conflicting styles like “photorealistic cartoon”.
Over-complicating: Too many elements can confuse the AI. Focus on the most important aspects.
Ignoring composition: Without composition guidance, the AI may not frame the image as you envision.

​
Tips for better results

Start broad, then refine: Begin with a general concept, then add specific details in subsequent generations.
Use reference images: Upload reference images to guide style and composition when available.
Experiment with models: Different models interpret prompts differently. Try multiple models for the same prompt.
Iterate and learn: Each generation teaches you what works. Keep notes on successful prompt patterns.

​
Prompt templates

Use these templates as starting points:

​
Portrait template

[Subject] portrait, [style], [lighting], [mood], [background description], [quality modifiers]

​
Landscape template

[Location] landscape, [time of day], [weather], [composition], [color palette], [style], [quality]

​
Product template

[Product] on [surface], [lighting setup], [background], [style], [angle], [quality modifiers]

Model Selection

Learn which models work best for different prompt types.

Reference Images

Use reference images to enhance your prompts.

Settings Guide

Understand how settings affect prompt interpretation.

Best Practices

More tips for getting great results.