Text to image AI generator

Five images in five different media, each generated from a text prompt on Morphic

A prompt in, a finished image out. Any style, any aspect ratio.

Every image model

One prompt box

Written as a prompt, returned as an image

From a sentence to an image in three steps

  1. 01

    Open Morphic

    Sign up and start creating on a free-flowing, infinite visual canvas.

  2. 02

    Write the prompt

    Name the subject, the style, the light, and the framing. Set your aspect ratio before you generate.

  3. 03

    Generate and refine

    Read what came back, change one element, run it again. Two or three passes usually lands it.

Anatomy of an image prompt

Four cues carry most of the result. Name the subject, the style you want it rendered in, where the light comes from, and how the shot is framed. Everything else is fine-tuning.
SubjectStyleLightFraming
SubjectA brass diving helmet on a workbenchStyleoil painting, visible brushworkLightsingle window light from the leftFramingtight three-quarter framing
SubjectAn old border collie asleep on a porchStyledocumentary photographLightlow afternoon sun, long shadowsFramingshallow depth of field, 50mm
SubjectA ramen shop on a wet side streetStylehand-painted anime backgroundLightwarm signage glow against blue duskFramingwide establishing shot

The same subject, two prompts

Both images came from the same model in the same run. The only thing that changed was how much the prompt actually said. On the left, "a dog on a porch". On the right, "an old border collie asleep on a weathered porch, low afternoon sun, shallow depth of field, 50mm". Drag the handle to see what the extra detail buys you.

An old border collie asleep on a weathered porch in low afternoon sun, shot with shallow depth of field on a 50mm lensFull prompt
A generic image of a dog on a porch generated from a four-word promptFour-word prompt

Use cases

Concepts before anyone commissions art

Put a look in front of the room while the idea is still cheap to change. Write the scene four ways, generate all four, and argue about pictures instead of adjectives.

Concepts before anyone commissions art

Editorial and blog artwork

Articles need images that no stock library happens to have. Describe the exact scene the piece is about and generate it in the ratio your layout expects.

Editorial and blog artwork

Product and campaign visuals

Test a colourway, a setting, and a mood in an afternoon. Keep the lighting and lens wording fixed so the set reads as one campaign rather than six unrelated shots.

Product and campaign visuals

Backgrounds, textures, and plates

Generate the surface, sky, or backdrop a composite is missing, at the ratio you need it. Cheaper than a reshoot and faster than hunting through a library.

Backgrounds, textures, and plates

Prompts worth stealing

Six prompts that show what each cue is doing. Open any of them, swap the subject, and keep the rest to hold the look.
A lighthouse in heavy dawn fog rendered in cold blue light on a long lens

Fogbound lighthouse

A lighthouse in heavy fog at dawn, cold blue light, long lens, muted palette

Edit prompt
A still life of citrus fruit and a ceramic jug lit by a single softbox against a dark backdrop

Studio still life

Still life of citrus and a ceramic jug, single softbox from the right, dark backdrop

Edit prompt
A mountain valley painted in loose ink and wash on pale paper with open negative space

Ink and wash landscape

Mountain valley in ink and wash, loose brushwork, pale paper texture, lots of empty space

Edit prompt
An isometric flat-colour illustration of a woodworking workshop with soft shading

Isometric workshop

Isometric illustration of a woodworking shop, flat colour, soft ambient occlusion

Edit prompt
A portrait of a weathered fisherman in flat overcast daylight with visible film grain

Film grain portrait

Portrait of a fisherman, overcast daylight, 35mm film grain, weathered detail

Edit prompt
A travel poster of a coastal town built from torn paper collage in a three-colour palette

Paper collage poster

Travel poster of a coastal town, torn paper collage, three-colour palette, bold type space

Edit prompt

All on Morphic

Everything the prompt box opens up

Simple pricing

Get started for free today, with the option to upgrade or cancel anytime.

Basic

$9/ month
billed as $0 per year

1100 monthly credits

1 user only

All models

Workflows

Standard

$24/ month
billed as $0 per year

3625 monthly credits

1 user only

All models

Workflows

Pro

$45/ month
billed as $0 per year

6350 shared monthly credits

1 user

+ up to 4 more at extra cost

All models

Workflows

Pro Max

$170/ month
billed as $0 per year

24650 shared monthly credits

1 user

+ up to 9 more at extra cost

All models

Workflows

Enterprise

For higher limits

Custom

pricing and billing terms

High-volume credits
Custom seat limits
All models
Workflows
Pricing Gradient

Free

For playing around

$0

forever free

Up to 20 credits
1 user only
Limited models
Workflows

FAQs

How does text to image AI work?
The model reads your sentence and builds a picture from noise, refining it step by step until the result lines up with the words you gave it. It is not searching a library or stitching stock photos together, so every image comes out new. What you write is the whole instruction set, which is why a specific prompt lands closer than a vague one.
Is text to image free to use?
Yes. Morphic has a free tier, so you can write a prompt and generate an image without paying anything up front. Paid plans raise how much you can generate and cover commercial use, but the text to image tool itself is open to you from the first prompt.
Do I need to sign up to generate an image?
You can browse the examples on this page without an account. To generate your own image you need a free account, which takes under a minute and no card. That is what gives you somewhere to save your generations and come back to them later.
What makes a good text to image prompt?
Name four things: the subject, the style, the lighting, and the framing. "A dog" gives the model almost nothing to work with. "An old border collie asleep on a wooden porch, late afternoon light, shallow depth of field, shot on 50mm" gives it a subject, a mood, a light source, and a lens. Add or swap one element at a time so you can see what each change does.
Can I choose which model generates the image?
Yes. Nano Banana 2, Flux 2 Pro, Seedream 5 Pro, Ideogram 4.0, Krea 2, and ChatGPT Images 2.0 all sit in the same workspace, and you can switch between them on the same prompt. They have real differences: some hold text and typography better, some render skin and natural light better, some are faster. Running one prompt through two of them is the quickest way to see which suits your subject.
What aspect ratios and resolutions can I generate?
Square, portrait, landscape, and wide formats are all available, so you can generate for a feed post, a phone wallpaper, or a banner without cropping afterwards. Set the ratio before you generate rather than trimming later, because the model composes the shot to fill the frame you give it.
Can I keep a style consistent across several images?
Yes. Keep the style, lighting, and lens wording identical and change only the subject line, and the set holds together. For a longer series, save the prompt as a Workflow so the same recipe runs every time instead of being retyped and drifting.

You might also like