AI Compare · Test kit
Make a picture that follows a detailed brief
One paragraph with countable details: a tram with exactly three passengers, a woman in a green coat with a white dog, four houses with red roofs, an evening sky and no text. Score the first image against each detail.
Prices for United States, before sales tax Data release 2026.10.16-3, Oct 16, 2026 Rules 2.1.0
Run the test
- Download the file for the AI.
- Copy the prompt.
Paste the brief from the file (the part between the lines) into the AI exactly as written, and score the first image it gives you.
- Open the AI in another tab, in your own account, start a fresh chat, attach the file and paste the prompt.
- Note what you used: the plan, the visible model or mode, whether web search was on, how many attempts it took and anything that went wrong.
- Score the answer with the checks below. Run the same kit in each AI you are choosing between.
Scorecard
Critical checks are marked. One failed critical check matters more than a nice style.
- A yellow tram on rails, on a stone bridge over a riverCritical
- Exactly three passengers visible in the tramCritical
- A woman in a green coat walking a white dog on a red lead, the same way as the tramCritical
- Exactly four houses with red roofs behind the bridgeCritical
- No text, sign, logo or watermark anywhereCritical
- The second house from the left has a round window
- Orange sky near the horizon, dark blue at the top, lit street lamps
- Landscape, in a flat illustration style
To compare without knowing which AI made which, ask someone else to save the pictures as A, B and C before you score them, and to tell you which is which afterwards.
Reference answer (open after you score)
There is no reference picture: score against the brief, and count rather than estimate. Counts and positions (three passengers, four houses, the second house's round window, the dog on a lead going the same way) are where pictures most often drift from a brief; a lovely image with five houses fails that check. Do not regenerate until you like the result: score the first image, and record the attempts if it failed to appear.
Kit version 1.0.0, files frozen on Oct 13, 2026. Original fixture by What the tech, released under CC BY 4.0. Fictional people, companies and figures. A passing score shows how one answer did on this kit; it does not prove the tool is generally better.