# Seedream 4.0 on the anchor ladder v3

Blind pairwise test on the photo-anchored quality scale. Ratings collected 2026-09-06 to 2026-09-11.

## Files

- `votes.csv` — every judgement, one row each: scene, category, the two sides compared, rater, rater role, and the choice on each question (the winning side's id, `tie`, or empty if unanswered).
- `results.json` — scores with 95% intervals, pairwise contrasts, results by category and the study design.
- `scenes.csv` — every scene: category, aspect ratio, prompt and, where a real photograph is the reference, its source (own photography or the Unsplash author and original).
- Images — separate archive, linked from the model page.

## Questions

- `prompt_adherence` — Prompt adherence: Which image follows the prompt better?
- `believability` — Believability: Which is more believable as a real photograph?
- `aesthetics` — Aesthetics: Which is more attractive?

## Participants

- `seedream-4-0` — Seedream 4.0
- `qwen-image-2` — Qwen Image 2.0
- `gpt-image-2` — GPT Image 2
- `flux-2-klein` — FLUX.2 Klein
- `real-photograph` — Real photograph

## Read before reusing

- The model ignores the requested output size: Seedream returned 2352 px on the long edge where 2048 was asked. Output was downscaled to match the ladder; it was never upscaled.
- No seed control: Seedream does not accept a seed, so the three generations per scene are independent but not reproducible. This measures average ability rather than a best pick.
- Prompt adherence is biased against photography, which is why the absolute score is built on believability alone.
- Tie share is 28.4%, higher than the 25.4% of the original ladder study. This session pairs two models that are genuinely indistinguishable, where “equal” is an honest answer.

## Licence

Data — CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Images — All rights reserved; provided for verification only.

## How to cite

> Everypixel research (2026). Seedream 4.0 on the anchor ladder v3 [Data set]. Ratings collected 6–11 September 2026.
