Fashion

Fashion boards, runway, regional clothing styles

8 tasks · 11 models tested · 88 results

Haute couture show

image
google gemini-2.5-flash-image
9.5/10 6.2 s
google gemini-2.5-flash-image
Cost < 0.01 $
Resolution 1344 x 768
Time 6.2 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.5
Review
The image perfectly respects all elements of the prompt, featuring an impressive avant-garde dress and a consistent minimalist setting. The technical quality is excellent, displaying sharpness and lighting worthy of a real high-fashion runway. The composition is balanced, and the aesthetic is highly sophisticated.
google gemini-3-pro-image-preview
9.5/10 27.7 s
google gemini-3-pro-image-preview
Cost < 0.01 $
Resolution 2752 x 1536
Time 27.7 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.5
Review
The image perfectly adheres to all prompt constraints: the haute couture style is evident through the avant-garde dress, and the minimalist setting is ideally rendered. The technical quality is excellent, featuring highly realistic lighting and texture management.
google imagen-4.0-fast-generate-001
9.5/10 4.7 s
google imagen-4.0-fast-generate-001
Cost 0.02 $
Resolution 1408 x 768
Time 4.7 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.5
Review
The image perfectly adheres to all elements of the prompt: the model, the avant-garde dress, and the minimalist runway are all present. The technical quality is excellent, featuring a photorealistic rendering and very professional lighting management. The composition is well-balanced, reinforcing the haute couture aesthetic of the scene.
google imagen-4.0-generate-001
9.5/10 7.2 s
google imagen-4.0-generate-001
Cost 0.04 $
Resolution 1408 x 768
Time 7.2 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.5
Review
The image is of exceptional technical quality, featuring a striking photorealistic rendering. The composition perfectly adheres to the conventions of fashion photography, and the dress design is truly avant-garde, thus responding with precision to the entirety of the prompt.
google imagen-4.0-ultra-generate-001
9.5/10 12.4 s
google imagen-4.0-ultra-generate-001
Cost 0.08 $
Resolution 1408 x 768
Time 12.4 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.5
Review
The image is technically flawless, with sharpness and light management worthy of professional fashion photography. The composition perfectly adheres to the requested minimalism, and the avant-garde dress demonstrates great creativity. The prompt is fully respected without any visible artifacts.
openai chatgpt-image-latest
9.5/10 44.8 s
openai chatgpt-image-latest
Cost 0.21 $
Resolution 1536 x 1024
Time 44.8 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.5
Review
The image perfectly adheres to all elements of the prompt, featuring a striking avant-garde dress and a very clean, minimalist runway. The technical quality is excellent, offering sharpness and light management worthy of a real haute couture fashion show. The composition is balanced and reinforces the high-end feel of the scene.
segmind ideogram-3
9.5/10 13.4 s
segmind ideogram-3
Cost 0.04 $
Resolution 1344 x 768
Time 13.4 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.5
Review
The image is technically flawless, featuring exceptional sharpness and highly professional lighting management. The composition perfectly adheres to the requested minimalist aesthetic, emphasizing the avant-garde structure of the dress. Prompt fidelity is absolute, capturing the very essence of haute couture without any superfluous artifice.
segmind seedream-4.5
9.5/10 10.8 s
segmind seedream-4.5
Cost 0.04 $
Resolution 2560 x 1472
Time 10.8 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.5
Review
The image perfectly adheres to all elements of the prompt: the model, the avant-garde dress, and the minimalist runway are all present. The technical quality is excellent, featuring highly realistic lighting and texture management. The composition is elegant and typical of professional fashion photography.
segmind seedream-v5-lite
9.1/10 26.9 s
segmind seedream-v5-lite
Cost 0.04 $
Resolution 2848 x 1600
Time 26.9 s
Matania Judgment
Quality
8
Composition
8
Creativity
7
Text accuracy
10
Fidelity
10
Overall
9.13
Review
The image adheres perfectly to the prompt, capturing the essence of a haute couture runway with an avant-garde dress on a minimalist catwalk. The technical quality is high, featuring excellent texture rendering and a balanced composition that highlights the subject. The absence of text in the image results in a maximum textual accuracy score by default.
xai grok-imagine-image
9.5/10 6.6 s
xai grok-imagine-image
Cost 0.02 $
Resolution 2752 x 1504
Time 6.6 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.5
Review
The image perfectly adheres to all prompt constraints, including the avant-garde style and the minimalist podium. The technical quality is excellent, featuring highly realistic lighting management and fabric textures. The composition is balanced and typical of professional fashion photography.
xai grok-imagine-image-pro
9.5/10 13.9 s
xai grok-imagine-image-pro
Cost 0.07 $
Resolution 2816 x 1536
Time 13.9 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.5
Review
The image perfectly adheres to all prompt instructions, capturing the essence of haute couture with a highly detailed, avant-garde dress. The composition is elegant, and the technical rendering is exceptionally sharp, with no visible artifacts. The minimalism of the runway enhances the luxurious and professional feel of the scene.

Accessories board

image
google gemini-2.5-flash-image
6.4/10 5.4 s
google gemini-2.5-flash-image
Cost < 0.01 $
Resolution 1344 x 768
Time 5.4 s
Matania Judgment
Quality
9
Composition
9
Creativity
7
Text accuracy
10
Fidelity
4
Overall
6.38
Review
The image is technically superb, featuring photorealistic rendering and an impeccable luxury aesthetic. However, the model fails significantly on the prompt's quantitative constraint: there are not 12 accessories present, but only a handful of objects, which drastically impacts the faithfulness score.
google gemini-3-pro-image-preview
6.4/10 25.4 s
google gemini-3-pro-image-preview
Cost < 0.01 $
Resolution 2752 x 1536
Time 25.4 s
Matania Judgment
Quality
9
Composition
9
Creativity
7
Text accuracy
10
Fidelity
4
Overall
6.38
Review
The image is technically superb, featuring a realistic rendering and impeccable luxury aesthetics. However, the model fails heavily on the quantitative constraint: there are not 12 accessories, but far fewer, which significantly penalizes the faithfulness score.
google imagen-4.0-fast-generate-001
6.4/10 4.4 s
google imagen-4.0-fast-generate-001
Cost 0.02 $
Resolution 1408 x 768
Time 4.4 s
Matania Judgment
Quality
9
Composition
9
Creativity
7
Text accuracy
10
Fidelity
4
Overall
6.38
Review
The image is technically superb, featuring a realistic rendering and a very successful luxury aesthetic. However, the model fails heavily on the prompt's quantitative constraint: it does not generate 12 accessories, but a significantly lower amount, which heavily penalizes the faithfulness score.
google imagen-4.0-generate-001
6.4/10 9.2 s
google imagen-4.0-generate-001
Cost 0.04 $
Resolution 1408 x 768
Time 9.2 s
Matania Judgment
Quality
9
Composition
9
Creativity
7
Text accuracy
10
Fidelity
4
Overall
6.38
Review
The visual aesthetics and technical quality are excellent, featuring a realistic and luxurious rendering of the marble and objects. However, the model fails heavily on the quantitative constraint: there are not 12 accessories, but only about 6 or 7, which significantly penalizes the fidelity score.
google imagen-4.0-ultra-generate-001
6.4/10 11.1 s
google imagen-4.0-ultra-generate-001
Cost 0.08 $
Resolution 1408 x 768
Time 11.1 s
Matania Judgment
Quality
9
Composition
9
Creativity
7
Text accuracy
10
Fidelity
4
Overall
6.38
Review
The aesthetics and technical image quality are excellent, achieving a highly successful luxury look. However, the model fails heavily on the quantitative constraint: the image contains far fewer than the 12 accessories requested in the prompt. This failure to follow numerical instructions severely impacts the faithfulness score.
openai chatgpt-image-latest
6.4/10 50.9 s
openai chatgpt-image-latest
Cost 0.21 $
Resolution 1536 x 1024
Time 50.9 s
Matania Judgment
Quality
9
Composition
9
Creativity
7
Text accuracy
10
Fidelity
4
Overall
6.38
Review
The image is technically superb, featuring a highly successful luxury aesthetic and a realistic rendering of the marble. However, the model fails heavily on the quantitative constraint: there are only about 7 to 8 visible accessories instead of the 12 explicitly requested, which significantly penalizes the faithfulness score.
segmind ideogram-3
6.9/10 13.1 s
segmind ideogram-3
Cost 0.04 $
Resolution 1344 x 768
Time 13.1 s
Matania Judgment
Quality
9
Composition
9
Creativity
7
Text accuracy
10
Fidelity
5
Overall
6.88
Review
The image is technically superb, featuring a highly successful luxury aesthetic and excellent light management on the marble. However, the model fails heavily on the quantitative constraint: there are not 12 accessories, but rather about ten less distinct objects, which significantly penalizes prompt fidelity.
segmind seedream-4.5
6.3/10 24.1 s
segmind seedream-4.5
Cost 0.04 $
Resolution 2560 x 1472
Time 24.1 s
Matania Judgment
Quality
8
Composition
9
Creativity
7
Text accuracy
10
Fidelity
4
Overall
6.25
Review
The visual aesthetics and technical quality are excellent, featuring highly successful rendering of marble and lighting. However, the model fails significantly on prompt adherence: it generates far fewer than the requested 12 accessories, which heavily penalizes the final score despite the image's beauty.
segmind seedream-v5-lite
6.3/10 38.0 s
segmind seedream-v5-lite
Cost 0.04 $
Resolution 2848 x 1600
Time 38.0 s
Matania Judgment
Quality
8
Composition
9
Creativity
7
Text accuracy
10
Fidelity
4
Overall
6.25
Review
The visual aesthetics are excellent, featuring great sharpness and an elegant composition on marble. However, the model fails heavily on the quantity constraint: there are not 12 accessories present, which causes the faithfulness score to plummet. While the image quality is undeniable, strict adherence to quantitative instructions is absent.
xai grok-imagine-image
6.4/10 6.9 s
xai grok-imagine-image
Cost 0.02 $
Resolution 2752 x 1504
Time 6.9 s
Matania Judgment
Quality
9
Composition
9
Creativity
7
Text accuracy
10
Fidelity
4
Overall
6.38
Review
The aesthetics and technical image quality are excellent, featuring a luxurious and realistic rendering of the marble and objects. However, the model fails significantly on the quantity constraint: there are not 12 accessories, but only about ten visible objects, which heavily penalizes prompt adherence.
xai grok-imagine-image-pro
5.9/10 13.7 s
xai grok-imagine-image-pro
Cost 0.07 $
Resolution 2816 x 1536
Time 13.7 s
Matania Judgment
Quality
9
Composition
9
Creativity
7
Text accuracy
10
Fidelity
3
Overall
5.88
Review
The image is technically superb, featuring a luxury aesthetic and realistic marble rendering. However, it fails heavily on prompt adherence: the model generated only about 6 to 7 accessories instead of the 12 explicitly requested. This failure to respect the quantitative constraint is a major fidelity error.

Traditional costumes

image
google gemini-2.5-flash-image
5.0/10 6.9 s
google gemini-2.5-flash-image
Cost < 0.01 $
Resolution 1344 x 768
Time 6.9 s
Matania Judgment
Quality
7
Composition
8
Creativity
7
Text accuracy
2
Fidelity
4
Overall
5
Review
The image fails the fidelity criterion because it does not display 8 distinct costumes, and the captions consist of incoherent and illegible text (gibberish). Although the visual aesthetics and composition are of good quality, the failure to respect the requested count and the lack of textual accuracy heavily penalize the score.
google gemini-3-pro-image-preview
5.0/10 26.5 s
google gemini-3-pro-image-preview
Cost < 0.01 $
Resolution 2752 x 1536
Time 26.5 s
Matania Judgment
Quality
7
Composition
8
Creativity
7
Text accuracy
2
Fidelity
4
Overall
5
Review
The image is aesthetically pleasing with high visual quality, but it fails significantly on textual and quantitative instructions. The model fails to generate legible or coherent captions (very low text_accuracy), and the number of costumes does not exactly match the request for eight distinct and clearly identified items (mediocre fidelity).
google imagen-4.0-fast-generate-001
4.9/10 6.3 s
google imagen-4.0-fast-generate-001
Cost 0.02 $
Resolution 1408 x 768
Time 6.3 s
Matania Judgment
Quality
7
Composition
8
Creativity
7
Text accuracy
1
Fidelity
4
Overall
4.88
Review
The image presents a beautiful visual aesthetic and a mood board-style composition, but it fails heavily on textual and quantitative constraints. The model failed to generate exactly 8 distinct costumes clearly and, most importantly, the text/captions are either completely illegible or nonexistent (character hallucinations), which directly violates the prompt.
google imagen-4.0-generate-001
5.0/10 11.1 s
google imagen-4.0-generate-001
Cost 0.04 $
Resolution 1408 x 768
Time 11.1 s
Matania Judgment
Quality
7
Composition
8
Creativity
7
Text accuracy
2
Fidelity
4
Overall
5
Review
The image fails on the crucial criteria of fidelity and textual accuracy: although it presents a costume board, the exact count of 8 is not respected, and the captions consist of incoherent text (gibberish). While the visual quality and composition are good, the model's inability to generate legible text and adhere to a precise count significantly penalizes the evaluation.
google imagen-4.0-ultra-generate-001
5.0/10 14.4 s
google imagen-4.0-ultra-generate-001
Cost 0.08 $
Resolution 1408 x 768
Time 14.4 s
Matania Judgment
Quality
8
Composition
7
Creativity
7
Text accuracy
2
Fidelity
4
Overall
5
Review
The image is visually high-quality, but it fails significantly on textual and quantitative constraints. The model failed to respect the exact count of 8 costumes, and the generated text is unreadable gibberish (textual hallucination), which is critical for a "captioned plate." Consequently, fidelity is very low despite a pleasing overall aesthetic.
openai chatgpt-image-latest
5.3/10 49.8 s
openai chatgpt-image-latest
Cost 0.22 $
Resolution 1536 x 1024
Time 49.8 s
Matania Judgment
Quality
8
Composition
9
Creativity
7
Text accuracy
2
Fidelity
4
Overall
5.25
Review
The image is aesthetically very successful, featuring excellent visual quality and a balanced composition. However, the model fails heavily on fidelity and textual accuracy: the text is unreadable gibberish (typical hallucinations of image models), and the exact number of costumes does not strictly match the prompt's requirement for 8 distinct, captioned elements.
segmind ideogram-3
6.0/10 15.5 s
segmind ideogram-3
Cost 0.04 $
Resolution 1344 x 768
Time 15.5 s
Matania Judgment
Quality
8
Composition
9
Creativity
7
Text accuracy
4
Fidelity
5
Overall
6
Review
The image is aesthetically very successful, boasting excellent visual quality and a balanced, concept art-style composition. However, the model fails on prompt adherence, as it only presents 4 outfits instead of the requested 8, and text precision is poor, resulting in illegible or incoherent captions.
segmind seedream-4.5
4.5/10 18.5 s
segmind seedream-4.5
Cost 0.04 $
Resolution 2560 x 1472
Time 18.5 s
Matania Judgment
Quality
7
Composition
6
Creativity
5
Text accuracy
2
Fidelity
4
Overall
4.5
Review
The model fails on the primary constraints of quantity and precision: it does not present 8 distinct costumes, and the captions consist of illegible and incoherent text. Although the visual aesthetics are correct, prompt adherence is very low due to the failure to respect the requested number of elements and the inability to generate intelligible text.
segmind seedream-v5-lite
4.0/10 47.3 s
segmind seedream-v5-lite
Cost 0.04 $
Resolution 2848 x 1600
Time 47.3 s
Matania Judgment
Quality
7
Composition
6
Creativity
5
Text accuracy
2
Fidelity
3
Overall
4
Review
The model fails significantly on prompt adherence: it does not generate 8 distinct outfits, and the requested caption consists of incoherent text (gibberish). Although the visual aesthetics are correct, the textual precision is almost non-existent, and the 'reference sheet' structure is not respected.
xai grok-imagine-image
5.0/10 7.4 s
xai grok-imagine-image
Cost 0.02 $
Resolution 2752 x 1504
Time 7.4 s
Matania Judgment
Quality
7
Composition
8
Creativity
7
Text accuracy
2
Fidelity
4
Overall
5
Review
The model fails on the primary constraints of quantity and textual precision: it does not present 8 distinct costumes, and the captions consist of incoherent text (gibberish). Although the visual aesthetics and texture quality are successful, the failure to respect the required number of elements and the inability to generate legible text heavily penalize fidelity and accuracy.
xai grok-imagine-image-pro
5.0/10 16.5 s
xai grok-imagine-image-pro
Cost 0.07 $
Resolution 2816 x 1536
Time 16.5 s
Matania Judgment
Quality
8
Composition
7
Creativity
7
Text accuracy
2
Fidelity
4
Overall
5
Review
The image fails to quantitatively adhere to the prompt: it does not present 8 distinct costumes, but rather a confused composition. Furthermore, fidelity is heavily impacted by the model's inability to generate legible and coherent captions, producing gibberish instead of actual costume names.

Fashion evolution

image
google gemini-2.5-flash-image
6.0/10 5.9 s
google gemini-2.5-flash-image
Cost < 0.01 $
Resolution 1344 x 768
Time 5.9 s
Matania Judgment
Quality
7
Composition
8
Creativity
7
Text accuracy
2
Fidelity
6
Overall
6
Review
The image presents a beautiful visual aesthetic and a balanced composition illustrating the evolution of styles. However, fidelity is penalized by the model's inability to generate legible and coherent text, rendering the concept of a 'timeline' purely visual rather than informative. The text consists of abstract and illegible characters, which is a major flaw for a task requiring a chronology.
google gemini-3-pro-image-preview
5.0/10 28.3 s
google gemini-3-pro-image-preview
Cost < 0.01 $
Resolution 2752 x 1536
Time 28.3 s
Matania Judgment
Quality
8
Composition
7
Creativity
7
Text accuracy
2
Fidelity
4
Overall
5
Review
The image fails the primary fidelity criterion because it does not present a structured timeline across 10 distinct decades, but rather a collage of styles. Furthermore, the generated text is completely incoherent and illegible, which seriously undermines the informative function requested by the prompt.
google imagen-4.0-fast-generate-001
6.0/10 3.8 s
google imagen-4.0-fast-generate-001
Cost 0.02 $
Resolution 1408 x 768
Time 3.8 s
Matania Judgment
Quality
7
Composition
8
Creativity
7
Text accuracy
2
Fidelity
6
Overall
6
Review
The image presents a pleasing visual aesthetic and a good separation of styles by era. However, the text is completely incoherent and illegible, which seriously undermines its function as a "timeline." Furthermore, although stylistic evolution is suggested, strict adherence to the requested 10 decades is difficult to verify due to the textual chaos.
google imagen-4.0-generate-001
5.0/10 13.5 s
google imagen-4.0-generate-001
Cost 0.04 $
Resolution 1408 x 768
Time 13.5 s
Matania Judgment
Quality
8
Composition
7
Creativity
7
Text accuracy
2
Fidelity
4
Overall
5
Review
The image fails in terms of fidelity because it does not present a structured timeline spanning 10 decades, but rather a montage of stylized portraits without any clear temporal distinction. The generated text is incoherent and illegible, which constitutes a major failure for a 'timeline' type prompt. The aesthetic quality is good, however, even though the concept of chronology is entirely absent.
google imagen-4.0-ultra-generate-001
5.0/10 31.2 s
google imagen-4.0-ultra-generate-001
Cost 0.08 $
Resolution 1408 x 768
Time 31.2 s
Matania Judgment
Quality
8
Composition
7
Creativity
7
Text accuracy
2
Fidelity
4
Overall
5
Review
The image is aesthetically pleasing with high visual quality, but it fails in terms of fidelity and textual accuracy. The model fails to generate a structured timeline, and the text/numbers are completely illegible or incoherent, which is a major flaw for an infographic/timeline-style request.
openai chatgpt-image-latest
6.0/10 50.3 s
openai chatgpt-image-latest
Cost 0.22 $
Resolution 1536 x 1024
Time 50.3 s
Matania Judgment
Quality
8
Composition
7
Creativity
7
Text accuracy
2
Fidelity
6
Overall
6
Review
The image is aesthetically pleasing and respects the visual evolution of fashion, but it fails on an informative level. Fidelity is penalized because the "timeline" structure is very imprecise and the text is completely illegible or incoherent (textual hallucinations), which is a major flaw for an infographic-style prompt.
segmind ideogram-3
9.1/10 14.6 s
segmind ideogram-3
Cost 0.04 $
Resolution 1344 x 768
Time 14.6 s
Matania Judgment
Quality
8
Composition
9
Creativity
7
Text accuracy
9
Fidelity
10
Overall
9.13
Review
The image follows the instructions perfectly, presenting a clear and aesthetic timeline covering the requested ten decades. The visual quality is excellent, featuring legible typography and a coherent stylistic progression for each era. The composition is well-balanced, and the technical execution is very clean.
segmind seedream-4.5
4.8/10 31.8 s
segmind seedream-4.5
Cost 0.04 $
Resolution 2560 x 1472
Time 31.8 s
Matania Judgment
Quality
7
Composition
6
Creativity
7
Text accuracy
2
Fidelity
4
Overall
4.75
Review
The image fails conceptually because it presents a collection of disparate portraits rather than a structured timeline. Fidelity is low as the '10 decades' constraint and the 'timeline' structure are not respected, and the generated text is completely incoherent (textual hallucinations).
segmind seedream-v5-lite
4.5/10 43.6 s
segmind seedream-v5-lite
Cost 0.04 $
Resolution 2848 x 1600
Time 43.6 s
Matania Judgment
Quality
7
Composition
6
Creativity
5
Text accuracy
2
Fidelity
4
Overall
4.5
Review
The image fails in terms of fidelity because it fails to structure a truly coherent timeline spanning 10 decades. Although the aesthetics are acceptable, the text is completely incoherent and illegible, making it impossible to understand the historical evolution. The composition resembles a fashion collage more than the ordered chronology that was requested.
xai grok-imagine-image
6.0/10 9.3 s
xai grok-imagine-image
Cost 0.02 $
Resolution 2752 x 1504
Time 9.3 s
Matania Judgment
Quality
7
Composition
8
Creativity
7
Text accuracy
2
Fidelity
6
Overall
6
Review
The image is aesthetically pleasing, featuring high rendering quality and a balanced composition that evokes a sense of progression. However, fidelity is compromised by the AI's inability to generate a legible and accurate timeline: the text is gibberish, and the precise count of 10 decades is not structurally respected. Precise text and chronology tasks remain a major weakness here.
xai grok-imagine-image-pro
6.0/10 28.0 s
xai grok-imagine-image-pro
Cost 0.07 $
Resolution 2816 x 1536
Time 28.0 s
Matania Judgment
Quality
8
Composition
7
Creativity
7
Text accuracy
2
Fidelity
6
Overall
6
Review
The image is aesthetically pleasing with good rendering quality, but it fails on textual accuracy, displaying the incoherent text typical of image generation models. Although the chronological aspect is visually suggested, the model fails to ensure a clear and exact distinction between the 10 requested decades, which impacts its fidelity to a precise 'timeline' structure.

Street style

image
google gemini-2.5-flash-image
9.4/10 8.1 s
google gemini-2.5-flash-image
Cost < 0.01 $
Resolution 1344 x 768
Time 8.1 s
Matania Judgment
Quality
9
Composition
8
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.38
Review
The image perfectly adheres to all prompt constraints, including the exact number of people (6) and the requested urban style. The technical quality is excellent, featuring a photorealistic rendering and a natural handling of city lighting. The composition is well-balanced, aesthetically and modernly capturing the essence of "street style."
google gemini-3-pro-image-preview
9.4/10 26.0 s
google gemini-3-pro-image-preview
Cost < 0.01 $
Resolution 2752 x 1536
Time 26.0 s
Matania Judgment
Quality
9
Composition
8
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.38
Review
The image perfectly adheres to all prompt constraints, including the exact number of people and the street-style aesthetic. The technical quality is excellent, featuring a realistic photographic render, masterful control of urban lighting, and a balanced composition.
google imagen-4.0-fast-generate-001
9.4/10 7.0 s
google imagen-4.0-fast-generate-001
Cost 0.02 $
Resolution 1408 x 768
Time 7.0 s
Matania Judgment
Quality
9
Composition
8
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.38
Review
The image perfectly adheres to all prompt constraints, including the precise number of people (6) and the street-style aesthetic. The technical quality is excellent, featuring realistic rendering of clothing textures and a natural handling of urban lighting. The composition is well-balanced and effectively captures the essence of city fashion.
google imagen-4.0-generate-001
9.5/10 13.0 s
google imagen-4.0-generate-001
Cost 0.04 $
Resolution 1408 x 768
Time 13.0 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.5
Review
The image perfectly adheres to all prompt constraints, including the exact number of people (6) and the desired urban style. The technical quality is excellent, featuring realistic lighting management and impressive sharpness. The composition is well-balanced and perfectly captures the requested 'street style' aesthetic.
google imagen-4.0-ultra-generate-001
9.5/10 12.6 s
google imagen-4.0-ultra-generate-001
Cost 0.08 $
Resolution 1408 x 768
Time 12.6 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.5
Review
The image is technically flawless, featuring high-quality sharpness and texture rendering. The model perfectly adheres to all prompt constraints, including the exact number of people, clothing style, and urban setting. The composition is well-balanced and captures the essence of "street style" in a highly aesthetic manner.
openai chatgpt-image-latest
9.5/10 50.1 s
openai chatgpt-image-latest
Cost 0.21 $
Resolution 1536 x 1024
Time 50.1 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.5
Review
The image perfectly adheres to all prompt constraints, including the exact number of people (6) and the requested urban style. The technical quality is excellent, featuring a realistic photographic render, natural light management, and a balanced composition typical of street fashion photography.
segmind ideogram-3
9.5/10 12.5 s
segmind ideogram-3
Cost 0.04 $
Resolution 1344 x 768
Time 12.5 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.5
Review
The image perfectly adheres to all prompt constraints, including the exact number of people (6) and the requested urban style. The technical quality is excellent, featuring photorealistic rendering and very natural lighting. The composition is well-balanced and perfectly captures a high-end 'street style' aesthetic.
segmind seedream-4.5
9.3/10 16.0 s
segmind seedream-4.5
Cost 0.04 $
Resolution 2560 x 1472
Time 16.0 s
Matania Judgment
Quality
8
Composition
9
Creativity
7
Text accuracy
10
Fidelity
10
Overall
9.25
Review
The image perfectly adheres to all prompt constraints, specifically the exact number of people and the urban style. The technical quality is high, featuring beautiful lighting management and a balanced composition that effectively highlights the clothing style. The aesthetic is consistent with the expectations of a street fashion photograph.
segmind seedream-v5-lite
9.1/10 31.2 s
segmind seedream-v5-lite
Cost 0.04 $
Resolution 2848 x 1600
Time 31.2 s
Matania Judgment
Quality
8
Composition
8
Creativity
7
Text accuracy
10
Fidelity
10
Overall
9.13
Review
The image perfectly adheres to all prompt constraints, including the exact number of people and the urban style. The technical quality is high, featuring beautiful light management and a balanced composition. The 'street style' aesthetic is rendered in a very realistic and professional manner.
xai grok-imagine-image
9.3/10 7.1 s
xai grok-imagine-image
Cost 0.02 $
Resolution 2752 x 1504
Time 7.1 s
Matania Judgment
Quality
9
Composition
8
Creativity
7
Text accuracy
10
Fidelity
10
Overall
9.25
Review
The image adheres perfectly to the prompt, precisely including the 6 people and the requested street-style aesthetic. The technical quality is excellent, featuring a realistic photographic render and a balanced composition. The subject is well-anchored in an urban setting consistent with contemporary fashion.
xai grok-imagine-image-pro
9.3/10 15.7 s
xai grok-imagine-image-pro
Cost 0.07 $
Resolution 2816 x 1536
Time 15.7 s
Matania Judgment
Quality
9
Composition
8
Creativity
7
Text accuracy
10
Fidelity
10
Overall
9.25
Review
The image adheres perfectly to the prompt, including the exact number of people and the requested urban style. The technical quality is excellent, featuring a photorealistic rendering and very natural lighting. The composition is well-balanced, although the fashion style is quite conventional.

Shoes board

image
google gemini-2.5-flash-image
4.4/10 6.8 s
google gemini-2.5-flash-image
Cost < 0.01 $
Resolution 1344 x 768
Time 6.8 s
Matania Judgment
Quality
7
Composition
8
Creativity
6
Text accuracy
2
Fidelity
3
Overall
4.38
Review
The image fails to adhere to major structural constraints: instead of presenting a 4x4 grid of 16 shoes, it shows a more random composition. Furthermore, the text is nearly illegible or incoherent, which completely compromises the request for captions and precision. Although the visual aesthetics are acceptable, prompt fidelity is very low.
google gemini-3-pro-image-preview
5.0/10 27.7 s
google gemini-3-pro-image-preview
Cost < 0.01 $
Resolution 2752 x 1536
Time 27.7 s
Matania Judgment
Quality
7
Composition
8
Creativity
7
Text accuracy
2
Fidelity
4
Overall
5
Review
The model fails on the primary constraints of quantity and structure: it does not generate a 4x4 grid of 16 shoes, but rather a smaller number of objects in an irregular layout. Furthermore, the text generation (captions) is extremely degraded and illegible, which contradicts the request for captioned plates.
google imagen-4.0-fast-generate-001
4.4/10 3.9 s
google imagen-4.0-fast-generate-001
Cost 0.02 $
Resolution 1408 x 768
Time 3.9 s
Matania Judgment
Quality
7
Composition
8
Creativity
6
Text accuracy
2
Fidelity
3
Overall
4.38
Review
The image fails heavily on prompt adherence: it does not respect the requested 4x4 grid structure and does not contain 16 types of shoes. Furthermore, the generated text is completely illegible and incoherent, rendering the 'captioned board' function non-existent.
google imagen-4.0-generate-001
4.4/10 7.6 s
google imagen-4.0-generate-001
Cost 0.04 $
Resolution 1408 x 768
Time 7.6 s
Matania Judgment
Quality
7
Composition
8
Creativity
6
Text accuracy
2
Fidelity
3
Overall
4.38
Review
The model fails heavily on prompt adherence: instead of producing a 4x4 grid of 16 shoes, it generates a much more disordered and smaller arrangement. Furthermore, the text is total gibberish, meaning the requested caption is non-existent and unusable. While the visual quality of the objects is acceptable, the requested logical structure is absent.
google imagen-4.0-ultra-generate-001
4.8/10 15.0 s
google imagen-4.0-ultra-generate-001
Cost 0.08 $
Resolution 1408 x 768
Time 15.0 s
Matania Judgment
Quality
8
Composition
9
Creativity
7
Text accuracy
2
Fidelity
3
Overall
4.75
Review
The image is aesthetically very successful, featuring excellent visual quality and a clean grid composition. However, it fails heavily on structural constraints: it does not contain 16 shoes (only 9 are visible in a 3x3 grid) and the text is completely illegible or incoherent, directly contradicting the captioning and 4x4 format instructions.
openai chatgpt-image-latest
4.9/10 43.9 s
openai chatgpt-image-latest
Cost 0.22 $
Resolution 1536 x 1024
Time 43.9 s
Matania Judgment
Quality
7
Composition
8
Creativity
6
Text accuracy
2
Fidelity
4
Overall
4.88
Review
The image fails on both major prompt constraints: it does not present a 4x4 grid (it shows an irregular layout) and does not contain 16 types of shoes. Furthermore, the text is completely illegible and incoherent (textual hallucinations), rendering the 'labeled board' function entirely non-functional despite the correct visual aesthetics.
segmind ideogram-3
5.0/10 13.9 s
segmind ideogram-3
Cost 0.04 $
Resolution 1344 x 768
Time 13.9 s
Matania Judgment
Quality
8
Composition
9
Creativity
7
Text accuracy
4
Fidelity
3
Overall
5
Review
The model fails heavily on the primary constraint of fidelity: instead of producing a 4x4 grid of 16 shoes, it generates a more free-form and less structured composition. Although the visual quality and aesthetics are excellent, the text is illegible or incoherent, and the number of objects fails to adhere to the strict quantitative instruction.
segmind seedream-4.5
4.9/10 26.5 s
segmind seedream-4.5
Cost 0.04 $
Resolution 2560 x 1472
Time 26.5 s
Matania Judgment
Quality
7
Composition
8
Creativity
6
Text accuracy
2
Fidelity
4
Overall
4.88
Review
The model fails on the major constraints of quantity and structure: instead of producing a 4x4 grid of 16 shoes, it generates fewer items arranged randomly. Furthermore, the requested legend is almost non-existent or composed of illegible text (gibberish), which heavily penalizes its fidelity and textual accuracy.
segmind seedream-v5-lite
3.8/10 54.0 s
segmind seedream-v5-lite
Cost 0.04 $
Resolution 2848 x 1600
Time 54.0 s
Matania Judgment
Quality
6
Composition
5
Creativity
5
Text accuracy
2
Fidelity
3
Overall
3.75
Review
The model fails heavily on prompt adherence: instead of producing a 4x4 grid of 16 shoes, it generates a cluttered composition. Textual accuracy is almost non-existent (unreadable or incoherent text), and the "reference sheet" structure is not respected.
xai grok-imagine-image
4.9/10 8.0 s
xai grok-imagine-image
Cost 0.02 $
Resolution 2752 x 1504
Time 8.0 s
Matania Judgment
Quality
7
Composition
8
Creativity
6
Text accuracy
2
Fidelity
4
Overall
4.88
Review
The image fails to adhere to the requested quantitative structure, failing to present an exact 4x4 grid of 16 shoes. While the aesthetics are of good quality, the text is completely incoherent or illegible, which heavily penalizes faithfulness and textual precision.
xai grok-imagine-image-pro
4.4/10 17.3 s
xai grok-imagine-image-pro
Cost 0.07 $
Resolution 2816 x 1536
Time 17.3 s
Matania Judgment
Quality
7
Composition
8
Creativity
6
Text accuracy
2
Fidelity
3
Overall
4.38
Review
The image fails significantly on prompt adherence: it respects neither the 4x4 grid (only about 12 elements are present) nor the exact count of 16 shoes. Furthermore, textual precision is almost nonexistent, with the captions consisting of illegible characters and gibberish, which is a major flaw for a labeled reference sheet.

Fabric board

image
google gemini-2.5-flash-image
4.9/10 7.5 s
google gemini-2.5-flash-image
Cost < 0.01 $
Resolution 1344 x 768
Time 7.5 s
Matania Judgment
Quality
8
Composition
7
Creativity
6
Text accuracy
2
Fidelity
4
Overall
4.88
Review
The image exhibits excellent visual quality and a beautiful diversity of textures, but it fails heavily on structural constraints. The prompt precisely required 12 captioned samples, yet the model produced an imprecise number of samples with illegible and incoherent text, which severely penalizes fidelity and textual accuracy.
google gemini-3-pro-image-preview
4.9/10 26.0 s
google gemini-3-pro-image-preview
Cost < 0.01 $
Resolution 2752 x 1536
Time 26.0 s
Matania Judgment
Quality
8
Composition
7
Creativity
6
Text accuracy
2
Fidelity
4
Overall
4.88
Review
The image exhibits excellent visual quality and highly realistic textile textures. However, it fails heavily on prompt fidelity: the number of samples is incorrect, and the generated text is illegible and incoherent, failing to fulfill the "caption" instruction. The model prioritizes image aesthetics at the expense of the requested structural and textual constraints.
google imagen-4.0-fast-generate-001
4.9/10 4.3 s
google imagen-4.0-fast-generate-001
Cost 0.02 $
Resolution 1408 x 768
Time 4.3 s
Matania Judgment
Quality
7
Composition
8
Creativity
6
Text accuracy
2
Fidelity
4
Overall
4.88
Review
The image presents a beautiful visual aesthetic and realistic textures, but fails to meet the structural constraints of the prompt. The sample count is not respected (far more than 12) and the generated text is complete gibberish, which heavily penalizes fidelity and textual accuracy.
google imagen-4.0-generate-001
4.9/10 8.0 s
google imagen-4.0-generate-001
Cost 0.04 $
Resolution 1408 x 768
Time 8.0 s
Matania Judgment
Quality
8
Composition
7
Creativity
6
Text accuracy
2
Fidelity
4
Overall
4.88
Review
The image exhibits excellent visual quality and realistic textile texture, but it fails to meet the structural constraints of the prompt. The number of samples does not match the requested 12, and the generated text is incoherent or illegible, which heavily penalizes its fidelity and textual accuracy.
google imagen-4.0-ultra-generate-001
4.9/10 13.1 s
google imagen-4.0-ultra-generate-001
Cost 0.08 $
Resolution 1408 x 768
Time 13.1 s
Matania Judgment
Quality
8
Composition
7
Creativity
6
Text accuracy
2
Fidelity
4
Overall
4.88
Review
The image is technically high-quality with realistic textures, but it fails on the prompt's major structural constraints. The sample count is incorrect (approximately 6 instead of 12), and the caption text is completely incoherent or illegible, which heavily penalizes both faithfulness and textual accuracy.
openai chatgpt-image-latest
5.3/10 45.6 s
openai chatgpt-image-latest
Cost 0.22 $
Resolution 1536 x 1024
Time 45.6 s
Matania Judgment
Quality
8
Composition
9
Creativity
7
Text accuracy
2
Fidelity
4
Overall
5.25
Review
The image is aesthetically pleasing with realistic textures, but it fails in terms of prompt fidelity. The model does not respect the quantitative constraint of 12 samples and, most importantly, the captions consist of incoherent and illegible text. Textual accuracy is very low, which heavily penalizes the final score despite the excellent visual quality.
segmind ideogram-3
5.9/10 15.6 s
segmind ideogram-3
Cost 0.04 $
Resolution 1344 x 768
Time 15.6 s
Matania Judgment
Quality
8
Composition
8
Creativity
7
Text accuracy
4
Fidelity
5
Overall
5.88
Review
The image exhibits excellent visual quality and highly realistic textile textures. However, prompt adherence is insufficient as the model fails to generate exactly 12 distinct samples, and the captions consist of incoherent, unreadable gibberish, thereby failing to provide the requested informative aspect.
segmind seedream-4.5
4.9/10 12.8 s
segmind seedream-4.5
Cost 0.04 $
Resolution 2560 x 1472
Time 12.8 s
Matania Judgment
Quality
7
Composition
8
Creativity
6
Text accuracy
2
Fidelity
4
Overall
4.88
Review
The image fails significantly on prompt fidelity because it does not present 12 distinct samples, but rather a more global composition of textures. Furthermore, the captioning constraint is completely ignored, as the generated text is illegible and incoherent. However, the visual quality of the textures is satisfactory.
segmind seedream-v5-lite
4.9/10 49.3 s
segmind seedream-v5-lite
Cost 0.04 $
Resolution 2848 x 1600
Time 49.3 s
Matania Judgment
Quality
7
Composition
8
Creativity
6
Text accuracy
2
Fidelity
4
Overall
4.88
Review
The image presents a beautiful visual aesthetic and a good representation of textile textures. However, it fails on major structural constraints: the number of samples is not 12, and the generated text is completely illegible and incoherent (gibberish), which heavily penalizes textual fidelity and precision.
xai grok-imagine-image
4.9/10 9.2 s
xai grok-imagine-image
Cost 0.02 $
Resolution 2752 x 1504
Time 9.2 s
Matania Judgment
Quality
8
Composition
7
Creativity
6
Text accuracy
2
Fidelity
4
Overall
4.88
Review
The image demonstrates excellent rendering quality for textile textures, but fails heavily on structural constraints. The prompt precisely required 12 samples and a caption; however, the model generates a random number of textures accompanied by illegible, incoherent text (gibberish). Consequently, fidelity is very low despite the successful visual aesthetics.
xai grok-imagine-image-pro
4.9/10 17.9 s
xai grok-imagine-image-pro
Cost 0.07 $
Resolution 2816 x 1536
Time 17.9 s
Matania Judgment
Quality
8
Composition
7
Creativity
6
Text accuracy
2
Fidelity
4
Overall
4.88
Review
The image is visually aesthetic with good texture management, but it fails on major structural constraints. The number of samples does not adhere to the instruction of 12, and more importantly, the captions consist of incoherent and illegible text, which contradicts the 'captioned' requirement. Fidelity is therefore heavily impacted by the failure to respect both the count and textual readability.

Jewelry collection

image
google gemini-2.5-flash-image
6.3/10 6.6 s
google gemini-2.5-flash-image
Cost < 0.01 $
Resolution 1344 x 768
Time 6.6 s
Matania Judgment
Quality
9
Composition
8
Creativity
7
Text accuracy
10
Fidelity
4
Overall
6.25
Review
The image is technically excellent, featuring a luxurious rendering and well-executed dramatic lighting. However, prompt adherence is poor, as the model generated only a single piece of jewelry instead of the collection of 8 pieces explicitly requested.
google gemini-3-pro-image-preview
6.3/10 24.2 s
google gemini-3-pro-image-preview
Cost < 0.01 $
Resolution 2752 x 1536
Time 24.2 s
Matania Judgment
Quality
9
Composition
8
Creativity
7
Text accuracy
10
Fidelity
4
Overall
6.25
Review
The image is technically superb, featuring dramatic lighting and an undeniable luxury aesthetic. However, the model fails heavily on the prompt's quantitative constraint: it does not present a collection of 8 pieces, but rather a much more limited number of objects. This counting error severely impacts the faithfulness score, despite the high visual quality.
google imagen-4.0-fast-generate-001
7.5/10 4.0 s
google imagen-4.0-fast-generate-001
Cost 0.02 $
Resolution 1408 x 768
Time 4.0 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
6
Overall
7.5
Review
The image is technically superb, featuring dramatic lighting and high-quality rendering. However, the model fails on a major quantitative constraint: instead of presenting a collection of 8 pieces, it shows a different number of objects, which heavily penalizes the faithfulness score.
google imagen-4.0-generate-001
6.3/10 7.3 s
google imagen-4.0-generate-001
Cost 0.04 $
Resolution 1408 x 768
Time 7.3 s
Matania Judgment
Quality
9
Composition
8
Creativity
7
Text accuracy
10
Fidelity
4
Overall
6.25
Review
The image is technically superb, featuring dramatic lighting and an undeniable luxury aesthetic. However, it fails heavily on prompt adherence: the model presents a single central piece of jewelry rather than a "collection of 8 pieces." This failure to respect a major quantitative constraint severely impacts the final score.
google imagen-4.0-ultra-generate-001
7.5/10 10.1 s
google imagen-4.0-ultra-generate-001
Cost 0.08 $
Resolution 1408 x 768
Time 10.1 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
6
Overall
7.5
Review
The image is technically superb, featuring high-quality rendering, successful dramatic lighting, and an undeniable luxury aesthetic. However, prompt fidelity is insufficient because the model does not display exactly 8 pieces of jewelry as requested, which was an explicit quantitative constraint.
openai chatgpt-image-latest
6.4/10 44.4 s
openai chatgpt-image-latest
Cost 0.21 $
Resolution 1536 x 1024
Time 44.4 s
Matania Judgment
Quality
9
Composition
9
Creativity
7
Text accuracy
10
Fidelity
4
Overall
6.38
Review
The image is technically superb, featuring dramatic lighting and an undeniable luxury aesthetic. However, the model failed on a major quantitative constraint: there are not 8 pieces presented, but a significantly lower number, which heavily penalizes the faithfulness score.
segmind ideogram-3
7.5/10 14.5 s
segmind ideogram-3
Cost 0.04 $
Resolution 1344 x 768
Time 14.5 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
6
Overall
7.5
Review
The aesthetic and technical quality is exceptional, featuring a luxurious finish and perfectly mastered dramatic lighting. However, the model fails on a crucial quantitative constraint: the image displays a significantly different number of pieces than the 8 requested, which heavily impacts the fidelity score.
segmind seedream-4.5
6.3/10 17.5 s
segmind seedream-4.5
Cost 0.04 $
Resolution 2560 x 1472
Time 17.5 s
Matania Judgment
Quality
9
Composition
8
Creativity
7
Text accuracy
10
Fidelity
4
Overall
6.25
Review
The image is technically superb, featuring dramatic lighting and exemplary sharpness, but it fails heavily on the prompt's quantitative constraint. Instead of presenting a collection of 8 pieces as requested, the model shows a significantly lower number, which heavily penalizes the faithfulness score.
segmind seedream-v5-lite
5.9/10 40.3 s
segmind seedream-v5-lite
Cost 0.04 $
Resolution 2848 x 1600
Time 40.3 s
Matania Judgment
Quality
8
Composition
7
Creativity
6
Text accuracy
10
Fidelity
4
Overall
5.88
Review
The image is aesthetically pleasing, featuring high-quality rendering and effective dramatic lighting. However, prompt adherence is very poor, as the model fails to generate the requested 8 pieces, presenting far fewer objects and a less structured layout than what was expected for a "collection."
xai grok-imagine-image
6.3/10 6.9 s
xai grok-imagine-image
Cost 0.02 $
Resolution 2752 x 1504
Time 6.9 s
Matania Judgment
Quality
9
Composition
8
Creativity
7
Text accuracy
10
Fidelity
4
Overall
6.25
Review
The image is technically excellent, featuring a high-end aesthetic and very successful dramatic lighting. However, prompt adherence is poor because the model fails to respect the quantitative constraint: it displays significantly fewer than the 8 pieces requested. While the aesthetic quality is present, the specific instruction regarding quantity is ignored.
xai grok-imagine-image-pro
6.5/10 14.6 s
xai grok-imagine-image-pro
Cost 0.07 $
Resolution 2816 x 1536
Time 14.6 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
4
Overall
6.5
Review
The image is technically superb, featuring a luxurious render and perfectly mastered dramatic lighting. However, prompt fidelity is poor, as the model only generates a single piece of jewelry instead of the 8 pieces explicitly requested. Aesthetic quality does not compensate for this major failure to respect quantitative constraints.
Code