AA Image Edit

In plain words · Which AI edited pictures the way people preferred (Artificial Analysis)

What does it measure?

A ranking of how well an AI edits an existing picture as instructed. It works like the Arena Intelligence image edit ranking but uses a separate pool of votes from another organization.

Who evaluates it, and how?

In the Image Editing Arena of independent evaluator Artificial Analysis, two anonymous AIs’ edits of the same image with the same instruction are shown, and the choice is scored Elo-style.

Reading the score

This is a head-to-head record, not an exam score. There is no maximum; a 100-point gap means the leader is preferred about 64% of the time. When an AI has several quality settings, the setting with the most samples is shown, and AIs with few samples can swing widely. The scale differs from Arena Intelligence, so do not compare the numbers directly. Values in this table are currently about 900–1,200.