How 20 AI image models render: “A reunion of long-lost brother and sister, as depicted in Bangladeshi films. The brother is now a thug, he is overweight, with badly cut, long, curly hair. He wears a sleeveless T-shirt. A large knife peeks out of his trousers. His arms open wide to hug his sister. He stinks. So a few flies hover around his head. His sister, typically the film heroine, and therefore very attractive. But she's a cop wearing a cap and body-fitting khaki police dress. A gun in her holster, a baton by her belt, plus a walkie talkie in her one hand. Her arms are also open wide to hug the brother. Both are tearful. They must be placed on one side of the image. On the other side of the image, we place some products from the film that we want to sell: A circular locket that is open like a book, with childhood pictures of the brother and the sister on two sides. Below this open locket, we show a closed locket. Add text: Exclusive reunion lockets on sale.”
Same prompt, sent verbatim, at each model's default settings and a fixed seed. Judge: is the text exactly right, spelled correctly, well set? 41 community votes so far.
Community prompt by @sharierkhan · voting in progress — model names and win rates are revealed when this prompt retires from the rotation
Expected text: Exclusive reunion lockets on sale.
20 text-to-image models have rendered this typography prompt with the identical text at their default settings. Voting is in progress: the images are shown without attribution so votes stay blind, and the model behind each one, with its win rate, is revealed when the prompt retires from the rotation. 41 votes so far.
What voters look for: correct spelling, placement and legibility, and type that looks printed or carved rather than pasted on.
- Model 1Revealed when voting closesVoting in progress
- Model 2Revealed when voting closesVoting in progress
- Model 3Revealed when voting closesVoting in progress
- Model 4Revealed when voting closesVoting in progress
- Model 5Revealed when voting closesVoting in progress
- Model 6Revealed when voting closesVoting in progress
- Model 7Revealed when voting closesVoting in progress
- Model 8Revealed when voting closesVoting in progress
- Model 9Revealed when voting closesVoting in progress
- Model 10Revealed when voting closesVoting in progress
- Model 11Revealed when voting closesVoting in progress
- Model 12Revealed when voting closesVoting in progress
- Model 13Revealed when voting closesVoting in progress
- Model 14Revealed when voting closesVoting in progress
- Model 15Revealed when voting closesVoting in progress
- Model 16Revealed when voting closesVoting in progress
- Model 17Revealed when voting closesVoting in progress
- Model 18Revealed when voting closesVoting in progress
- Model 19Revealed when voting closesVoting in progress
- Model 20Revealed when voting closesVoting in progress
FAQ
- Why are the model names hidden?
- This prompt is still being voted on. The arena is blind: voters compare two images without knowing which model made them. Names and win rates appear here once the prompt retires from the rotation.
- Can I vote on this prompt?
- Yes — pairs from every live prompt are served on the arena vote page. You will not be shown a prompt you wrote yourself.
- Were the prompts adjusted per model?
- No. Every model received the exact same prompt, unmodified, at its default settings, with no negative prompt or style suffix.