Which AI image model follows a hard composition
This prompt is a counting and placement test: a blue book on the left, a glass of water with three ice cubes in the middle, a red apple on the right, a pencil balanced on the glass. We asked 36 models and checked every position. The grid includes the September 18 additions; each model page dates its own run.
Anvisha Pai, Co-founder & CEO, Voyager
What the side-by-side shows
GPT Image 2.5 Flare, Nano Banana Pro, FLUX.2 Pro and Grok Imagine 2.0 placed everything exactly, pencil balanced and all. Most of the roster is close. Recraft V3 put the pencil on the table and the apple behind the glass; Recraft V4.1 stood the pencil inside the glass; Ideogram 4.0 and Seedream 5.0 Lite lost the count of ice cubes. Ideogram 4.5, added on 22 October, placed the book and apple correctly but stood the pencil in the water instead of across the rim. Of the 22 October additions, FLUX 3 Image Klein, Seedream 5.0 Flash, Muse Image, MOSS-Image, Recraft V4.1 Flash and MAI-Image-2.5 Pro placed all five objects with three ice cubes; Krea 3 and Bria FIBO 2.0 drew four cubes, MAI-Image-2.5 four or five. Ideogram Vector places the book, apple and pencil correctly but adds more than three ice shapes.
- 1.GPT Image 2.5 Flare every object in its place, pencil balanced on the rim.
- 2.Nano Banana Pro exact placement with window light.
- 3.FLUX.2 Pro correct order and count, cleanly lit.
- 4.Grok Imagine 2.0 three cubes, pencil on the glass, apple on the right.
- 5.FLUX 3 Image Klein every object placed and counted, in under two seconds.
Every image model on this prompt, 35 pictures
One try each, nothing re-rolled or retouched, generated by us. Ranked models come first; the rest follow in family order. Open a model for the rest of its set.
›The prompt every model got
A glass of water on a wooden table with exactly three ice cubes in it, a yellow pencil resting across the rim of the glass, a blue book to the left of the glass and a red apple to the right, photographed from table height.
Exact counts and left-right placement of five named objects.
every object in its place, pencil balanced on the rim.
exact placement with window light.
correct order and count, cleanly lit.
three cubes, pencil on the glass, apple on the right.
every object placed and counted, in under two seconds.
The other prompts
- Which AI image model holds an art style
- Which AI image model handles a busy scene best
- Which AI image model draws an accurate chart
- Which AI image model edits a picture best
- Which AI image model does faces and hands best
- Which AI image model makes the best food photo
- Which AI image model makes the best infographic
- Which AI image model makes the best logo
- Which AI image model makes the best product shot on white
- Which AI image model renders text best
- Which AI image model makes the best wide banner
Frequently asked questions
Which AI image generator follows prompts most precisely?
On a five-object placement test, GPT Image 2.5 Flare, Nano Banana Pro, FLUX.2 Pro and Grok Imagine 2.0 got every position right. The Recraft models were the loosest.
Voyager
Run supported models inside Voyager
Voyager is an agent for creative work that runs supported models. Brief it, and it picks the model for the job, makes the thing and hands back something you can edit. Create an account to get started.