Model leaderboard

Every model runs the same Glif tests. Scores come from 2,042 blind votes by 10 Glif team members, plus 1,279 AI judge scores, each counting as half a vote. Each kind of task has its own board.

Score out of 100: how often people would pick this model's output over an average model's on the same tests. 50 is average.

How we score

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.

Bars show each score’s 90% range. Overlapping bars are close to a tie.

Hide vote counts

254 team votes on 7 tests. A model ranks once it has 20 votes’ worth of ratings here.

  1. 1

    Seedream 5.0 Pro

    90% range 65–78
    Wins 18 / Losses 6 / Ties 16 · 40 team votes · 11 AI judge scores

    71 out of 100, 90% range 65 to 78

  2. 2

    Ideogram V4.5Glif default

    90% range 58–78
    Wins 10 / Losses 7 / Ties 9 · 26 team votes · 5 AI judge scores

    69 out of 100, 90% range 58 to 78

  3. 3

    FLUX 3 Image

    90% range 60–77
    Wins 8 / Losses 5 / Ties 13 · 26 team votes · 8 AI judge scores

    69 out of 100, 90% range 60 to 77

  4. 4

    GPT Image 2.5Glif default

    90% range 60–74
    Wins 18 / Losses 9 / Ties 21 · 48 team votes · 11 AI judge scores

    68 out of 100, 90% range 60 to 74

  5. 5

    Nano Banana Pro

    90% range 52–71
    Wins 9 / Losses 3 / Ties 13 · 25 team votes · 9 AI judge scores

    62 out of 100, 90% range 52 to 71

  6. 6

    Muse ImageGlif default

    90% range 51–66
    Wins 14 / Losses 12 / Ties 16 · 42 team votes · 10 AI judge scores

    59 out of 100, 90% range 51 to 66

  7. 7

    Krea 3Glif default

    90% range 43–70
    Wins 7 / Losses 10 / Ties 9 · 26 team votes · 0 AI judge scores

    58 out of 100, 90% range 43 to 70

  8. 8

    Gemini 3.1 FlashGlif default

    90% range 44–63
    Wins 7 / Losses 7 / Ties 7 · 21 team votes · 9 AI judge scores

    54 out of 100, 90% range 44 to 63

  9. 9

    Seedream 5.0 Lite

    90% range 38–63
    Wins 9 / Losses 6 / Ties 4 · 19 team votes · 7 AI judge scores

    52 out of 100, 90% range 38 to 63

  10. 10

    Phota Image Editor

    90% range 41–64
    Wins 10 / Losses 10 / Ties 8 · 28 team votes · 8 AI judge scores

    52 out of 100, 90% range 41 to 64

  11. 11

    Grok Imagine Image

    90% range 26–42
    Wins 2 / Losses 8 / Ties 10 · 20 team votes · 7 AI judge scores

    33 out of 100, 90% range 26 to 42

  12. 12

    FLUX.2 Pro

    90% range 22–43
    Wins 2 / Losses 7 / Ties 10 · 19 team votes · 7 AI judge scores

    32 out of 100, 90% range 22 to 43

  13. 13

    Seedream V4.5

    90% range 15–30
    Wins 1 / Losses 9 / Ties 7 · 17 team votes · 7 AI judge scores

    22 out of 100, 90% range 15 to 30

  14. 14

    Z-Image Generator

    90% range 11–23
    Wins 0 / Losses 11 / Ties 7 · 18 team votes · 7 AI judge scores

    18 out of 100, 90% range 11 to 23

Not ranked yet (2)