Long and detailed prompt
Every model got the same prompt. Here's what 26 models made, best first.
Results
Won
GPT Image 2.5
Score 66 · Judge 8.2/10
#2 of 13
Seedream 5.0 Pro
Score 59 · 0 wins, 0 losses, 1 tie · Judge 8/10
#3 of 13
Nano Banana Pro
Score 59 · 1 win, 0 losses · Judge 7/10
From 6 blind votes by the Glif team and 13 AI judge scores. Scores are out of 100, and 50 is average for this test. See the full text-to-image leaderboard
How we score
Scores on this page come from this test alone and compare only the models that ran it. Leaderboard scores pool every test of the same kind.
Every model runs the same Glif tests. Team members see two outputs side by side without the model names and pick the one that better fits the test, or call it a tie. An AI judge also scores every output from 0 to 10.
A Bradley-Terry fit turns all those head-to-head results into one rating per model. Ties count as half a win, and each AI judge score counts as half a vote.
The score is the predicted chance people prefer this model's output to the average model's on the same board, so 50 is average. A board is one kind of task, like text-to-image. Scores only compare models on the same board.
The 90% range comes from resampling the votes and refitting many times. Overlapping ranges are close to a tie. A model ranks once it has 20 votes' worth of ratings on a board. Until then it shows no score, only how close it is.
What we look for
Adheres to a long and detailed prompt.
Prompt
“Studio photograph of a silver panther on the left, a giant beach ball in the middle and a tall man on the right. The panther has long fur and green horns, its claws are attached to garden hoses that end up in milkshakes, the left scene is a highway restaurant. The beachball has chrome and black sections and is in surrounded by water, the man is wearing hi-viz street fashion and is cosplaying as a techbro and is surrounded by multiple signs that 'we're so back'.”
Tests for
Prompt adherence, spatial reasoning