AI image generator: from text prompt to finished image

Write what you want to see, pick a model, choose a format. Visuatelier sends your prompt to the model you chose and returns the image to your library.

Open the studio

Different text-to-image models are good at different things. Ideogram and GPT Image 2 handle lettering on posters and packaging well, Recraft leans towards clean design and vector-like styles, Seedream 4.5 and FLUX Pro 1.1 Ultra produce large, detailed stills, and FLUX Schnell is the quick option for rough drafts. In Visuatelier you can try the same prompt on several of them without opening separate accounts.

Each image has a fixed credit price shown before you generate. If a generation fails or our safety check blocks the result, the credits are returned.

How it works

1
Describe the subject, the setting, the light and the style in one or two sentences.
2
Choose a model and an aspect ratio (1:1, 4:5, 3:2, 16:9 or 9:16).
3
Generate, compare, then download or remix the prompt on another model.

What you can make

Product and packaging mock-ups
Explore lighting, materials and backgrounds before a real shoot.
Posters and social visuals
Use a typography-strong model when the image has to carry words.
Concept art and mood boards
Sketch worlds, characters and colour palettes quickly.
Illustration styles
Watercolour, ink, flat vector, clay or 3D renders from the same idea.

Examples from our showcase

Showcase images were generated by Visuatelier. The prompt and the model used are shown under each one.

Neon rain · Made with Seedream 4.5Prompt: Cinematic close-up portrait of a fictional young woman with silver-streaked hair standing in Tokyo-style neon rain at night, magenta and cyan reflections on wet skin, shallow depth of field, anamorphic bokeh, 85mm film still, ultra detailed
Sky archipelago · Made with Seedream 4.5Prompt: Epic fantasy landscape of floating islands with waterfalls cascading into clouds, ancient stone temples, glowing turquoise lakes, dragons gliding in the distance, sunrise volumetric light, matte painting
Clay village · Made with Recraft V4Prompt: Miniature claymation village on a floating island, tiny houses with colorful roofs, windmills, little trees, tilt-shift look, handmade clay texture, bright playful palette
Summer fest poster · Made with Ideogram V3Prompt: Bold retro music festival poster with the large title text "SUMMER WAVES" and smaller text "VISUATELIER SOUND FEST", sunset gradient, palm silhouettes, groovy 70s typography, risograph texture
Matcha cloud · Made with Nano Banana 2.1 (Google)Prompt: Overhead shot of a matcha mille crepe cake with strawberries and edible flowers on a pastel pink table, soft natural light, food styling, vibrant fresh colors
Desert wanderer · Made with GPT Image 2Prompt: Editorial portrait of a fictional elderly desert nomad with deep weathered skin and an indigo headscarf, golden hour light, dust particles glowing in the air, Kodak Portra film look, intense eyes

Models for this tool

FLUX Schnell · 1 credits
The fast draft model. It is the only one the daily free credits can buy.
FLUX Dev · 3 credits
A slower FLUX model for stills that need more care than a draft.
FLUX Pro 1.1 · 5 credits
FLUX Pro 1.1 for campaign stills, jewelry, and packaging you might actually print.
FLUX Pro 1.1 Ultra · 3 credits
FLUX Pro 1.1 Ultra renders up to four megapixels with a natural, photographic finish.
Seedream 4.5 · 2 credits
ByteDance's Seedream 4.5 makes sharp, detailed stills at about four megapixels.
Nano Banana 2.1 (Google) · 3 credits
Google's Nano Banana 2.1 follows long prompts closely and keeps scenes coherent.
GPT Image 2 · 3 credits
OpenAI's GPT Image 2 is strong at legible text and instruction-following.
Grok Imagine · 1 credits
xAI's Grok Imagine makes quick, stylish stills at a low price.
Recraft V4 · 2 credits
Recraft V4 is built for design work: clean layouts, brand-ready styles, and vector-like art.
Ideogram V3 · 3 credits
Ideogram V3 handles typography and graphic composition better than most models.

Frequently asked questions

Which AI image models can I use?

The image catalogue currently lists FLUX Schnell, FLUX Dev, FLUX Pro 1.1, FLUX Pro 1.1 Ultra, Seedream 4.5, Nano Banana 2.1, GPT Image 2, Grok Imagine, Recraft V4 and Ideogram V3. Each model page shows its current price.

How much does one image cost?

Images are priced per output in credits, from 1 credit with FLUX Schnell. The exact price appears on the model card before you generate.

Is there a free option?

Every account receives 10 free credits per UTC day, which can be spent on FLUX Schnell. Other models use paid credits.

Can I write prompts in French or German?

Yes, you can type in any language. Most models follow English prompts most reliably, so if a result misses a detail, try rephrasing that detail in English.

Are the images labelled?

Yes. Images made in Visuatelier are labelled AI-generated in the library, in the gallery and in page metadata.