TotalApp Docs

Image Creator

Generate stunning AI images from text prompts with 62 style presets and multiple aspect ratios.

Overview

Image Creator is TotalApp's text-to-image generator. You describe an image in plain language and the AI generates it in seconds. The tool supports 62 curated style presets that steer the AI's aesthetic output — from photorealistic product shots to hand-painted watercolours — and a full range of aspect ratios covering every use case from social media posts to cinema-grade wide shots.

Generated images are full resolution and can be downloaded immediately. The image history panel shows your recent generations so you can return to any previous image without regenerating it.

Writing Effective Prompts

The quality of an AI-generated image is directly proportional to the quality of your prompt. A vague prompt produces a vague image. A detailed, structured prompt produces a precise, intentional image. Use this structure as a guide:

  1. Subject — who or what is the main focus? (e.g. "a golden retriever puppy", "a futuristic city skyline", "a ceramic coffee mug")
  2. Action or pose — what is the subject doing? (e.g. "sitting on a park bench", "at dawn from a rooftop", "steam rising from the top")
  3. Setting / environment — where is the scene? (e.g. "in a snowy forest", "urban street at night", "minimalist white studio")
  4. Lighting — how is the scene lit? (e.g. "golden hour sunlight", "dramatic side lighting", "soft diffused studio lighting")
  5. Mood / atmosphere — how should it feel? (e.g. "peaceful and serene", "tense and dramatic", "warm and inviting")
  6. Style reference — reference specific visual styles if not using a preset (e.g. "impressionist brushstrokes", "bauhaus design", "1950s magazine advertisement")

Negative Prompts

Use the Negative Prompt field to tell the AI what to exclude. Common exclusions: "blurry", "watermark", "extra limbs", "low quality", "text", "signature". Negative prompts are especially effective for avoiding common AI artefacts like malformed hands or distorted backgrounds.

Style Presets

Style presets steer the AI's visual direction without requiring you to describe the style in your prompt. Select a preset and your prompt is automatically combined with the style's visual DNA. You can always override specific aspects of the preset by mentioning them explicitly in your prompt.

The 62 presets are grouped into categories:

  • Photographic — Photorealistic, Portrait, Product Shot, Macro, Aerial
  • Illustration — Digital Art, Watercolor, Oil Painting, Ink Drawing, Pencil Sketch
  • Animation — Anime, Cartoon, Pixar-Style, Comic Book, Flat Design
  • Architectural — Interior Design, Exterior Render, Blueprint, Isometric
  • Aesthetic — Cyberpunk, Vaporwave, Minimalist, Dark Fantasy, Cottagecore
  • Historical — Renaissance, Art Nouveau, Bauhaus, Pop Art, Expressionist

Aspect Ratios

Choose the right aspect ratio before generating — it determines the canvas shape and affects how the AI composes the scene.

Square (1:1)

Instagram posts, profile pictures, product thumbnails

Portrait (2:3)

Pinterest pins, book covers, story formats

Landscape (3:2)

Blog headers, print photos, email banners

Wide (16:9)

YouTube thumbnails, presentation slides, website hero images

Ultra-wide (21:9)

Cinema-style shots, desktop wallpapers, panoramic scenes

Variations and Remix

Once an image is generated, you have two ways to iterate on it without starting from scratch:

Variations

Click the Variations button on any generated image to produce 3–4 new versions using the same prompt and style but with different compositional choices. Variations are useful when the overall direction is right but you want to explore alternatives before committing.

Remix

Click Remix to open the Remix panel with the original image displayed alongside a prompt input. Describe the changes you want — "change the background to a forest", "make the lighting warmer and more golden", "add a coffee mug to the foreground" — and the AI generates a new image guided by both the original and your modification prompt. Remix preserves more of the original composition than a fresh generation.

Exporting Images

All generated images are available at full resolution for download. Use the following options in the image output panel:

  • Download button — saves the image as a PNG to your device's download folder at the original generation resolution.
  • Right-click → Copy image — copies the image to your clipboard for pasting directly into other applications (Figma, presentations, email drafts).
  • Right-click → Copy image address — copies the CDN URL of the image for embedding in web content.

Image History

The last 50 generated images are stored in your Image Creator history. Click the History button in the header to browse and re-download past generations. History is per-user and persists across sessions.

Frequently Asked Questions

What's the difference between Variations, Remix, and Style Transfer?
Variations regenerates 3–4 new takes on the same prompt and style, keeping the same overall direction but changing composition. Remix lets you describe a specific change ("make the background a forest") while keeping the rest of the original image intact. Style Transfer is a separate tool that applies a completely different visual style (e.g. turning a photo into a watercolor painting) while preserving the subject and composition. Use Variations to explore options, Remix to tweak a detail, and Style Transfer to change the artistic medium entirely.
Which aspect ratio should I use for social media posts?
Use Square (1:1) for Instagram feed posts and profile pictures, Portrait (2:3) for Pinterest pins and Instagram/Facebook Stories, and Wide (16:9) for YouTube thumbnails and Twitter/X header images. If you're unsure, Square is the safest default since it displays well across most platforms without cropping.
Can I combine a style preset with my own custom style description?
Yes. Selecting a style preset combines its visual DNA with your prompt automatically, but you can still mention specific style details in your own prompt text to override or refine parts of it — for example, selecting the "Photographic → Portrait" preset and adding "in black and white with dramatic shadows" in your prompt will apply both.
Why does my generated image not match my prompt exactly?
AI image generation is probabilistic, not deterministic — the same prompt can produce different results each time. If the output consistently misses the mark, make your prompt more specific using the six-part structure (Subject, Action/Pose, Setting, Lighting, Mood, Style Reference) and add a Negative Prompt to explicitly exclude unwanted elements like "blurry", "extra limbs", or "watermark". Using Remix on a close-but-not-quite result is often faster than starting over with a new prompt.