GenPix AI · Text to Image
Text to Image
Text to image is the simplest form of AI generation: words in, picture out, nothing else needed. It is the right tool when the image you want does not exist yet and nobody has photographed it.
Think of the prompt as a shot list rather than a wish. Name the subject, then the environment, then the light, then the medium. Adjectives like "beautiful" or "amazing" tell the model almost nothing; "backlit", "overcast", "macro", "grainy 35mm" tell it a great deal.
- 2 images free — no account, no card, no watermark
- Private by default — nothing you make is ever public unless you publish it yourself
- Never used for training — not your images, not your prompts
- What you make is yours — full resolution, commercial use, we take no cut
Made here, with the sentence that made it
Every picture here came out of this studio. Hover one to read the exact sentence that made it, press it to open the studio with that sentence already loaded.
-
A blade, neon and rainA female samurai in matte black armour with a glowing cyan katana, standing in a rain-soaked neon alley of a future city, holographic signs reflecting in puddles… -
A fae at the edge of the woodA woodland fae woman with translucent moth wings and a crown of ferns, freckled skin, mossy cloak, standing in a shaft of morning light in an ancient forest, drifting… -
Dust, sun and a long walkCinematic portrait of a woman desert nomad in layered sand-coloured wraps and a dust-covered face scarf, sun-cracked lips, carrying a worn staff, harsh afternoon light…
How to do it
- 1
Start with the subject in plain words.
- 2
Add the environment and the light — the two things that turn a drawing into a photograph.
- 3
Add a medium or camera hint at the end: oil painting, 35mm film, product render, pencil sketch.
Prompts that work
Press one to open the studio with it loaded — then change two words rather than face a blank box.
A glass greenhouse at night, lit from inside, rain on the panes, seen from across a wet lawnHand-drawn botanical plate of an invented fern, ink and watercolour on aged paperCinematic still: a red bus crossing an empty desert highway at golden hour, wide shot
Questions
- What resolution do I get?
- Up to 4096×4096 on the Max setting, and wide or tall variants of it. The usual render is smaller and costs 0.15 €; Max costs 0.23 €. Every image downloads at full size.
- Can I write in French?
- Yes. The model reads both, though English prompts are often a little more precise.
- Can I get several versions at once?
- Yes — ask for up to four images in one go and pick the one that works.