Domo AI · Text to image
Domo AI Text to Image
Type a description and get an image in about 10 seconds
Describe a subject, a place, a style and the light, and DomoAI turns your words into an image. Choose anime models for character art, or GPT Image and Nano Banana for posters, photoreal shots and 4K detail. This guide shows how to write descriptions that work.
≈ 10 sper image
30+models & styles
Up to 4Koutput
Prompt
A white tuk-tuk in a busy street market in Colombo, cinematic photography, golden hour sunlight, shallow depth of field, high detail
How it works
Image Generator overview →From words to image in 3 steps
- 01
Describe it
Write a prompt, or type a rough idea and let AI Optimize fill in the details.
- 02
Pick a model & ratio
Anime, fusion or a multi-purpose model, plus 1:1, 16:9 or 9:16.
- 03
Generate & refine
Use Vary to get 4 versions, or Re-gen. Then upscale, edit or animate the result.
Models
Which model to use
Five multi-purpose models from DomoAI’s image page, plus its anime and fusion families. Lineups change as new models launch.
| Model | Best for | Notes |
|---|---|---|
| GPT Image 2.5 · Flare | Fast everyday images, posters, thumbnails | Built for speed; 2K in Relax mode with up to 9 input images |
| GPT Image 2.5 · Sunburst | Fine detail, multi-step edits | Slower, holds detail across a sequence of edits |
| GPT Image 2 | Strong prompt following, short readable text | Good for ads and graphics with a headline |
| Nano Banana 2 | High-resolution, photoreal detail | 4K in Relax mode |
| Nano Banana Pro | Consistent characters | Up to 9 reference images |
| Anime S1–S10 | OCs, VTubers, anime key visuals | Soft portraits, idols, fantasy and cinematic anime looks |
| Fusion | Stylized portraits, game art | A blend of anime and realistic rendering |
Resolutions above what Relax mode covers use credits. All five multi-purpose models can both generate from text and edit an uploaded image.
Prompt formula
Six parts of a strong image prompt
Write them in this order. The model gives the most weight to what comes first, so lead with the subject.
- Subjectwho or what, with 2–3 details
- Environmentwhere, and when
- Stylephoto, anime, 3D, watercolor…
- Lightinggolden hour, neon, soft studio…
- Compositionclose-up, wide, low angle, lens
- Qualityhigh detail, sharp focus
Example, broken down
A white tuk-tuk in a busy street market in Colombo, cinematic photography, golden hour sunlight, shallow depth of field with bokeh, high detail, sharp focus
1 Subject2 Environment3 Style4 Lighting5 Composition6 Quality
Interactive
Image prompt builder
Fill each part and the builder writes the prompt, suggests a model and checks for common problems.
3 · Style
4 · Lighting
5 · Composition
Aspect ratio
Your prompt
Suggested model
Word bank
Prompt writing guide →Words that change the image
Click any phrase to copy it, then add it to your prompt.
Camera & lens
Composition
Style
Mood & color
Quality
Aspect ratios
Pick the shape for where it’s going
Choose the ratio before you generate. Cropping later cuts off parts of the image and wastes resolution.
1:1
16:9
9:16
| Ratio | Use it for |
|---|---|
| 1:1 | Feed posts, profile art, stickers, product shots |
| 16:9 | YouTube thumbnails, blog banners, desktop wallpapers |
| 9:16 | Stories, Reels, Shorts, phone wallpapers, and first frames for vertical video |
Getting readable text in images
- Use a model known for text, such as GPT Image 2 or 2.5
- Put the exact words in quotes: the headline “MOON DAY”
- Keep it short: 1–4 words works best
- Say where it goes: “bold title at the top”
- For long or exact copy, add text later in a design tool
Keeping a character consistent
- Create one clear “anchor” image first
- Reuse it as a reference (Nano Banana Pro takes up to 9)
- Describe each trait once: hair, outfit, colors
- Change only the scene between images
- Then animate it with Omni Reference
Troubleshooting
All limitations →Common problems and fixes
| Problem | Why | Fix |
|---|---|---|
| Odd hands or fingers | Complex hand poses are hard for every model | Hide or simplify hands (“hands in pockets”), or fix with the Image Editor |
| Face changes every time | No anchor for identity | Use a reference image and keep the description identical |
| Garbled text | Long text or a weak text model | Short quoted words with GPT Image, or add text afterwards |
| Cluttered, muddy result | Too many ideas in one prompt | One subject, one setting, one style; cut adjectives |
| Ignores part of the prompt | Key detail buried at the end | Move the important part to the front |
| Wrong style | Style word too vague (“nice art”) | Name the medium: “watercolor”, “anime key visual”, “35mm photo” |
Keep going
From image to video in one studio
Your generated image is a starting point. Use it in DomoAI’s other tools.
Use cases
What people make with text to image
Characters & OCs
Anime portraits and character sheets to animate later.
Social & thumbnails
Eye-catching 16:9 thumbnails and 9:16 story art.
Products & ads
Mockups, ad concepts and clean studio shots.
Concept art & banners
Scenes, worlds and blog headers.
What is Domo AI Text to Image?
DomoAI’s text-to-image tool turns a written description into an image. You describe the subject, setting, style and lighting, pick a model and aspect ratio, and get results in about 10 seconds.
Which models can I use?
DomoAI’s image page lists GPT Image 2, GPT Image 2.5 (Flare and Sunburst), Nano Banana 2 and Nano Banana Pro, plus anime models (S1–S10) and fusion models for stylized looks. Lineups change as new models launch.
How do I write a good text-to-image prompt?
Follow six parts in order: subject, environment, style, lighting, composition and quality. Lead with the subject, name a specific style or medium, and keep it to one main idea.
Can DomoAI put readable text in images?
Short text works best with GPT Image 2 or 2.5. Put the exact words in quotes and keep them to a few words. For long or exact copy, add the text afterwards in a design tool.
What resolution are the images?
It depends on the model. GPT Image 2.5 Flare covers 2K in Relax mode, and Nano Banana 2 covers 4K in Relax mode. You can also upscale up to 6K with the Image Upscaler.
How much does an image cost?
Costs depend on the model and resolution. DomoAI estimates Basic’s 600 credits make about 150 images, roughly 4 credits each. Standard and above include Relax mode on supported models.
How do I keep the same character across images?
Make one clear anchor image, reuse it as a reference (Nano Banana Pro accepts up to 9), describe each trait once, and change only the scene.
Can I use the images commercially?
DomoAI’s Terms say you keep ownership of what you create and may use it commercially if the use is lawful. Free outputs may be watermarked, so use a paid plan for client work.
Describe it. See it in seconds.
Write your first prompt and generate an image with free starter credits.