Side-by-side evidence

Compare

Seventeen verified runs: the two launch specimens plus a frozen three-prompt matrix across five hosted model endpoints or quality tiers.

Trial 01
ab-owl-identity-001-grok-trial-01

grok-imagine-image

Strong owl, navy enamel, brass, key, and art-deco fan adherence. Spectacles are round rather than half-moon. Exposed gears are plausible but not a demonstrably complete functional train.

Seed
not exposed
SHA-256
dea66004e613…
Policy
first party tool
Trial 02
ab-owl-identity-001-grok-trial-02

grok-imagine-image

Strong owl, navy enamel, brass key, and explicit art-deco geometry adherence. One amber eye is dominant. Spectacles are round rather than half-moon; visible gears mesh locally.

Seed
not exposed
SHA-256
7ef64f252650…
Policy
first party tool
Trial 03
r1-identity-grok-imagine

grok-imagine-image

Vision adherence score 8.6/10. Strong navy-and-brass clockwork owl, key, amber eye, art-deco fan, and material rendering. Spectacles read round rather than half-moon and the key is partly cropped.

Seed
not exposed
SHA-256
4e85a3d59021…
Policy
first party tool
Trial 04
r1-identity-minimax-image-01

image-01

Vision adherence score 6.4/10. Recognizable clockwork owl with brass details and a radial backdrop, but both eyes glow amber, the key is barely visible, red shoulder accents depart from the palette, and the fan geometry is less art-deco coherent.

Seed
not exposed
SHA-256
a094b098b6ef…
Policy
official api
Trial 05
r1-identity-openai-gpt-image-2-high

gpt-image-2-high

Vision adherence score 9.2/10. Excellent materials, key, navy/brass palette, coherent mechanisms, one amber eye, and art-deco framing. Eyewear is integrated and round rather than clearly half-moon.

Seed
not exposed
SHA-256
986a879709f4…
Policy
official api
Trial 06
r1-identity-openai-gpt-image-2-low

gpt-image-2-low

Vision adherence score 8.8/10. Strong navy/brass owl, visible key, one amber eye, coherent gears, and art-deco framing. Spectacles are round and the key is partly cropped.

Seed
not exposed
SHA-256
15844228e3fc…
Policy
official api
Trial 07
r1-identity-openai-gpt-image-2-medium

gpt-image-2-medium

Vision adherence score 9.5/10. Best overall identity adherence: prominent skeleton key, navy enamel, coherent brass gearwork, one amber eye, and strong art-deco geometry. Spectacles remain round rather than half-moon.

Seed
not exposed
SHA-256
9c4278c5d7ed…
Policy
official api
Trial 08
r1-spatial-grok-imagine

grok-imagine-image

Vision adherence score 10.0/10. Exact count and relations: 3 red cubes left, 2 glass spheres center, 1 brass pyramid behind them, 1 cream crane far right; 7 total.

Seed
not exposed
SHA-256
27eed3816d68…
Policy
first party tool
Trial 09
r1-spatial-minimax-image-01

image-01

Vision adherence score 3.0/10. Count failure: 5 red cubes, 3 glass spheres, 1 brass pyramid, and 1 cream crane; 10 total. Left-to-right category order is otherwise preserved.

Seed
not exposed
SHA-256
65fc2887173e…
Policy
official api
Trial 10
r1-spatial-openai-gpt-image-2-high

gpt-image-2-high

Vision adherence score 10.0/10. Exact count and relations: 3 red cubes left, 2 glass spheres center, 1 brass pyramid behind them, 1 cream crane far right; 7 total.

Seed
not exposed
SHA-256
53b58be88980…
Policy
official api
Trial 11
r1-spatial-openai-gpt-image-2-low

gpt-image-2-low

Vision adherence score 10.0/10. Exact count and relations: 3 red cubes left, 2 glass spheres center, 1 brass pyramid behind them, 1 cream crane far right; 7 total.

Seed
not exposed
SHA-256
12e002676189…
Policy
official api
Trial 12
r1-spatial-openai-gpt-image-2-medium

gpt-image-2-medium

Vision adherence score 10.0/10. Exact count and relations: 3 red cubes left, 2 glass spheres center, 1 brass pyramid behind them, 1 cream crane far right; 7 total.

Seed
not exposed
SHA-256
507a493f93b7…
Policy
official api
Trial 13
r1-typography-grok-imagine

grok-imagine-image

Vision adherence score 10.0/10. Exact two-line text with no additional generated words: MEASURE THE MAGIC / NOT THE MARKETING. Strong owl emblem, art-deco ornament, palette, legibility, and margins.

Seed
not exposed
SHA-256
e239625758d1…
Policy
first party tool
Trial 14
r1-typography-minimax-image-01

image-01

Vision adherence score 6.0/10. The two requested headline lines are spelled correctly, but an additional third line of pseudo-text appears below, violating the exact-text and no-other-words constraints.

Seed
not exposed
SHA-256
d194ba98f179…
Policy
official api
Trial 15
r1-typography-openai-gpt-image-2-high

gpt-image-2-high

Vision adherence score 9.6/10. Exact two-line text with no additional generated words, excellent owl emblem and art-deco construction. Slightly heavier ornamental density than requested but fully legible.

Seed
not exposed
SHA-256
2acc64c3fff7…
Policy
official api
Trial 16
r1-typography-openai-gpt-image-2-low

gpt-image-2-low

Vision adherence score 9.7/10. Exact two-line text with no additional generated words, strong clockwork owl emblem, art-deco framing, and clear hierarchy. Composition is denser than the Grok variant.

Seed
not exposed
SHA-256
0e6e41e117ff…
Policy
official api
Trial 17
r1-typography-openai-gpt-image-2-medium

gpt-image-2-medium

Vision adherence score 9.8/10. Exact two-line text with no additional generated words, excellent legibility, coherent mechanical owl emblem, restrained red accent, balanced margins, and strong art-deco geometry.

Seed
not exposed
SHA-256
ea42d0b9e548…
Policy
official api

Round-one signal

Model quality is task-shaped.

Grok Imagine and all three GPT Image 2 tiers rendered the exact seven-object spatial scene; MiniMax Image-01 produced ten objects. Grok and all three GPT tiers rendered the requested two-line poster text exactly; MiniMax added a third pseudo-text line. On Herman identity, GPT Image 2 Medium led the visual review at 9.5/10, followed by High at 9.2, Low at 8.8, Grok at 8.6, and MiniMax at 6.4. These are single unseeded trials, not statistical rankings.