Head to head
Esc

Models · slot A

to navigate to pick

Geometric Composition

Vote

A spatial-reasoning test. Each object has a precise relationship (inside, on top, behind, seen through the glass), so it measures whether a model follows explicit placement instructions and handles transparency and refraction rather than just approximating the scene.

Voters judged on
Spatial Accuracy Object Relationships Glass & Refraction Lighting Direction
The brief Every model got the same prompt
Prompt
A glass cube on a wooden table. Inside the cube is a small blue sphere. On top of the cube sits a red book. A green plant is behind the cube, partially visible through the glass. Soft window light from the left.
The leaderboard

Challenge rankings

60 models · ranked by blind vote

Through time: how it evolved
1
ImagineArt 1.5 (Preview)
score 28.9
-1
2
FLUX.2 [dev] Flash
score 27.7
0
3
FLUX.2 [dev] Turbo
score 27.7
0
4
Seedream 4.5
score 26.6
0
5
Nano Banana 2
score 25.3
0
6
GPT Image 1 Mini
score 24.0
0
7
Nano Banana Pro
score 23.4
1
8
GPT Image 1.5
score 23.4
0
9
Seedream 4.0
score 22.4
0
10
FLUX.2 [max]
score 21.7
0
11
Wan 2.5 (Preview)
score 21.5
0
12
FLUX.2 [flex]
score 21.5
0
13
Recraft V4
score 21.0
0
14
Nano Banana 2 Lite
score 21.0
0
15
FLUX.2 [dev]
score 20.8
0
16
Grok Imagine Image Pro
score 20.6
0
17
Z-Image Turbo
score 20.5
0
18
Nano Banana
score 20.0
0
19
Seedream 5.0 Lite
score 19.6
0
20
Imagen 4.0 Ultra Generate 001
score 18.4
0
21
FLUX.2 [pro]
score 18.4
0
22
Qwen Image 2512
score 18.0
0
23
Grok Imagine Image
score 17.8
0
24
GPT Image 1
score 17.3
0
25
Stable Diffusion 3.5 Large
score 16.4
0
26
Reve Image 1.0
score 15.9
0
27
Recraft V4 Pro
score 15.2
0
28
FLUX.1 Kontext [max]
score 15.2
0
29
FLUX.1 [schnell]
score 14.4
0
30
Recraft V4.1 Pro
score 12.8
0
31
Qwen Image
score 12.6
0
32
Bria FIBO
score 12.6
0
33
Qwen Image 2.0
score 12.2
0
34
Vidu Q2
score 11.2
0
35
Wan 2.6
score 10.4
0
36
Stable Diffusion 3.5 Large Turbo
score 10.4
0
37
FLUX.1 Kontext [dev]
score 10.4
0
38
DALL-E 3
score 9.8
0
39
FLUX.1 [schnell] FP8
score 8.1
0
40
Wan 2.7
score 7.9
0
41
Qwen Image 2.0 Pro
score 7.9
0
42
FLUX.2 [klein] 9B
score 7.9
0
43
FLUX.2 [klein] 4B
score 6.3
0
44
FLUX1.1 [pro] Ultra
score 6.2
0
45
FLUX1.1 [pro]
score 6.2
0
46
DALL-E 2
score 1.8
0
47
FLUX.1 Kontext [pro]
score 1.0
0
48
OmniGen v2
score 0.9
0
49
Stable Diffusion 3.5 Medium
score 0.8
0
50
LongCat-Image
score 0.8
0
51
Recraft V4.1
score 0.5
0
52
Wan 2.7 Pro
score 0.0
0
53
Qwen Image Max
score 0.0
0
54
FLUX.1 [dev]
score 0.0
0
HiDream I1 Full
awaiting votes
0
HiDream I1 Fast
awaiting votes
0
GPT Image 2
awaiting votes
0
FLUX.1 Krea [dev]
awaiting votes
0
Imagen 4.0 Generate 001
awaiting votes
0
Imagen 4.0 Fast Generate 001
awaiting votes
0
Head to head 4 matchups

Notable battles

The matchups worth a second look from this challenge's blind voting: the closest rivalries, the biggest upsets (a lower seed taking down a favorite), and the clashes at the top of the board. Each bar shows how the community split its votes.

Want to compare other models? Pick any two and see them go head to head.
The archive

Through time

How Geometric Composition has evolved, release by release.

2022 to 2026· 9 reigns· 60 models
The Reign Chain
9 reigns

How the crown changed hands

Every model that has held #1 on this challenge, in order. Click any era to travel through time.

  1. R01
    Inception
    DALL-E 2
    Apr 2022 Oct 2023
    reigned 1.5 yrs
  2. R02
    DALL-E 3
    Oct 2023 Aug 2024
    reigned 10 mo · overtook DALL-E 2
  3. R03
    FLUX.1 [schnell]
    Aug 2024 Oct 2024
    reigned 2 mo · overtook DALL-E 3
  4. R04
    Stable Diffusion 3.5 Large
    Oct 2024 Mar 2025
    reigned 5 mo · overtook FLUX.1 [schnell]
  5. R05
    GPT Image 1
    Mar 2025 Aug 2025
    reigned 5 mo · overtook Stable Diffusion 3.5 Large
  6. R06
    Imagen 4.0 Ultra Generate 001
    Aug 2025 Sep 2025
    reigned 1 mo · overtook GPT Image 1
  7. R07
    Seedream 4.0
    Sep 2025 Oct 2025
    reigned 1 mo · overtook Imagen 4.0 Ultra Generate 001
  8. R08
    GPT Image 1 Mini
    Oct 2025 Nov 2025
    reigned 1 mo · overtook Seedream 4.0
  9. R09
    Current
    ImagineArt 1.5 (Preview)
    Nov 2025 today
    reigned 8 mo · overtook GPT Image 1 Mini

Generation history

60 models · newest first
Show full history 55
  1. Recraft V4.1 3 takes
  2. GPT Image 2 3 takes
  3. Wan 2.7 3 takes
  4. 55 more models Apr 2022 – Apr 2026 Show