Art

Art styles and fine arts

4 tasks · 11 models tested · 44 results

Classical still life

image
google gemini-2.5-flash-image
9.0/10 8.8 s
google gemini-2.5-flash-image
Cost < 0.01 $
Resolution 1344 x 768
Time 8.8 s
Matania Judgment
Quality
8
Composition
9
Creativity
9
Text accuracy
10
Fidelity
9
Overall
9
Review
The image is visually stunning, featuring a balanced composition and remarkable technical mastery of texture and light. The model perfectly interpreted the abstract concept of a "never-before-seen style" by creating a hybrid and original aesthetic. The fidelity is excellent, as the work meets the demand for the extraordinary and the original without being constrained by limiting textual prompts.
google gemini-3-pro-image-preview
7.6/10 27.4 s
google gemini-3-pro-image-preview
Cost < 0.01 $
Resolution 2752 x 1536
Time 27.4 s
Matania Judgment
Quality
8
Composition
8
Creativity
7
Text accuracy
10
Fidelity
7
Overall
7.63
Review
The image is technically very successful, featuring a strong visual aesthetic and a balanced composition. However, the prompt was extremely vague ('never seen before style'), which makes the evaluation of creativity and faithfulness subjective, as the model generated a high-quality work without being able to truly prove the novelty of the requested style.
google imagen-4.0-fast-generate-001
7.6/10 4.3 s
google imagen-4.0-fast-generate-001
Cost 0.02 $
Resolution 1408 x 768
Time 4.3 s
Matania Judgment
Quality
8
Composition
8
Creativity
7
Text accuracy
10
Fidelity
7
Overall
7.63
Review
The image is technically successful, featuring a visually rich aesthetic and a balanced composition. However, because the prompt is extremely vague and subjective ("never-before-seen style"), the model produces a high-quality abstract work that lacks tangible evidence of a radical stylistic breakthrough; this limits the faithfulness score to a standard interpretation of abstract/surrealist art.
google imagen-4.0-generate-001
8.9/10 6.5 s
google imagen-4.0-generate-001
Cost 0.04 $
Resolution 1408 x 768
Time 6.5 s
Matania Judgment
Quality
8
Composition
8
Creativity
9
Text accuracy
10
Fidelity
9
Overall
8.88
Review
The image offers a visually rich and complex aesthetic that effectively meets the requirement for a "never seen before" style through the hybridization of textures and shapes. The technical quality is high, featuring excellent light management. The model perfectly interpreted the subjective and abstract aspects of the prompt.
google imagen-4.0-ultra-generate-001
6.6/10 13.7 s
google imagen-4.0-ultra-generate-001
Cost 0.08 $
Resolution 1408 x 768
Time 13.7 s
Matania Judgment
Quality
8
Composition
8
Creativity
7
Text accuracy
10
Fidelity
5
Overall
6.63
Review
The image is technically very successful, featuring a strong visual aesthetic and a balanced composition. However, the prompt is extremely vague and subjective ('never-before-seen style'), which makes assessing fidelity complex; while the model has produced a high-quality work, it is impossible to verify whether it truly adheres to a constraint of originality or if it relies on existing styles.
openai chatgpt-image-latest
8.9/10 51.2 s
openai chatgpt-image-latest
Cost 0.21 $
Resolution 1536 x 1024
Time 51.2 s
Matania Judgment
Quality
8
Composition
8
Creativity
9
Text accuracy
10
Fidelity
9
Overall
8.88
Review
The image perfectly addresses the subjective and open-ended nature of the prompt by offering a complex, abstract aesthetic that truly feels unprecedented. The technical quality is high, featuring remarkable management of textures and light, although the concept of a "never-before-seen style" is inherently difficult to measure objectively.
segmind ideogram-3
8.6/10 12.5 s
segmind ideogram-3
Cost 0.04 $
Resolution 1344 x 768
Time 12.5 s
Matania Judgment
Quality
8
Composition
8
Creativity
7
Text accuracy
10
Fidelity
9
Overall
8.63
Review
The image presents a visually stunning aesthetic with high technical mastery. The model interpreted the concept of a 'never-before-seen style' through a complex fusion of organic and abstract forms, perfectly adhering to the creative intent of the highly open-ended prompt.
segmind seedream-4.5
8.6/10 16.9 s
segmind seedream-4.5
Cost 0.04 $
Resolution 2560 x 1472
Time 16.9 s
Matania Judgment
Quality
8
Composition
8
Creativity
7
Text accuracy
10
Fidelity
9
Overall
8.63
Review
The image presents a strong visual aesthetic and high-quality technical execution with rich textures. Although the prompt is extremely subjective ('never-before-seen style'), the model succeeded in producing an abstract and complex work that aligns with it. The fidelity is excellent, as the image fulfills the intent of creating something extraordinary and original.
segmind seedream-v5-lite
8.6/10 39.1 s
segmind seedream-v5-lite
Cost 0.04 $
Resolution 2848 x 1600
Time 39.1 s
Matania Judgment
Quality
8
Composition
8
Creativity
7
Text accuracy
10
Fidelity
9
Overall
8.63
Review
The image is technically very well-executed, featuring impressive sharpness and a dreamlike aesthetic. It perfectly adheres to the prompt by offering an abstract and original vision that breaks away from conventional styles. The composition is balanced, and the "extraordinary" aspect is effectively captured through the use of colors and shapes.
xai grok-imagine-image
8.9/10 7.6 s
xai grok-imagine-image
Cost 0.02 $
Resolution 2752 x 1504
Time 7.6 s
Matania Judgment
Quality
8
Composition
8
Creativity
9
Text accuracy
10
Fidelity
9
Overall
8.88
Review
The image offers a complex and dreamlike aesthetic that perfectly meets the request for a "never-before-seen" style through the hybridization of organic textures and fractal structures. The technical quality is high, featuring impressive light management and detail. Prompt adherence is excellent, as the model interpreted the requested abstraction with great visual impact.
xai grok-imagine-image-pro
8.6/10 16.9 s
xai grok-imagine-image-pro
Cost 0.07 $
Resolution 2816 x 1536
Time 16.9 s
Matania Judgment
Quality
8
Composition
8
Creativity
7
Text accuracy
10
Fidelity
9
Overall
8.63
Review
The image presents a strong visual aesthetic and high-quality technical execution with interesting light management. Although the prompt is extremely vague ('never seen style'), the model succeeded in producing an abstract and complex work that meets the intention of being extraordinary. Fidelity is high, as the model interprets the subjectivity of the prompt through rich visual creation.

Old map

image
google gemini-2.5-flash-image
8.6/10 6.3 s
google gemini-2.5-flash-image
Cost < 0.01 $
Resolution 1344 x 768
Time 6.3 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
3
Fidelity
10
Overall
8.63
Review
The image perfectly respects all the thematic elements of the prompt (islands, monsters, compass rose) with a very successful parchment aesthetic. The visual quality and composition are excellent, creating an immersive atmosphere. However, as is often the case with image models, the text and annotations are illegible and nonsensical, which impacts textual precision.
google gemini-3-pro-image-preview
8.8/10 23.9 s
google gemini-3-pro-image-preview
Cost < 0.01 $
Resolution 2752 x 1536
Time 23.9 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
4
Fidelity
10
Overall
8.75
Review
The image perfectly adheres to all elements of the prompt (islands, monsters, compass rose) with a highly successful 17th-century aesthetic. The technical quality and composition are excellent, creating an immersive atmosphere. However, as is often the case with generative AI, the text and inscriptions on the map are illegible or consist of incoherent glyphs, which impacts textual accuracy.
google imagen-4.0-fast-generate-001
8.8/10 3.6 s
google imagen-4.0-fast-generate-001
Cost 0.02 $
Resolution 1408 x 768
Time 3.6 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
4
Fidelity
10
Overall
8.75
Review
The image perfectly adheres to all prompt elements (islands, monsters, compass rose) with a highly successful 17th-century aesthetic. The visual quality and composition are excellent, creating an immersive atmosphere. However, the text and inscriptions on the map exhibit typical AI artifacts (unreadable or incoherent characters), which impacts textual accuracy.
google imagen-4.0-generate-001
8.8/10 12.0 s
google imagen-4.0-generate-001
Cost 0.04 $
Resolution 1408 x 768
Time 12.0 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
4
Fidelity
10
Overall
8.75
Review
The image perfectly respects all elements of the prompt (islands, monsters, compass rose) with a very convincing 17th-century aesthetic. The visual quality and composition are excellent, reinforcing the authentic look of the map. Only the textual accuracy is lacking, with illegible characters and inconsistent glyphs, which is typical of image generation models.
google imagen-4.0-ultra-generate-001
8.8/10 11.7 s
google imagen-4.0-ultra-generate-001
Cost 0.08 $
Resolution 1408 x 768
Time 11.7 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
4
Fidelity
10
Overall
8.75
Review
The image perfectly respects all the semantic elements of the prompt (islands, monsters, compass rose) with a very convincing 17th-century aesthetic. The visual quality and composition are excellent, although the accuracy of text and typographic characters is poor, which is common for image generation models. Fidelity to the concept is absolute.
openai chatgpt-image-latest
8.8/10 52.2 s
openai chatgpt-image-latest
Cost 0.22 $
Resolution 1536 x 1024
Time 52.2 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
4
Fidelity
10
Overall
8.75
Review
The image perfectly respects all elements of the prompt (islands, monsters, compass rose) with a very successful 17th-century aesthetic. The technical quality and composition are excellent, creating an immersive atmosphere. However, as is often the case with generative AI, the text on the map is illegible and composed of incoherent characters, which impacts the textual accuracy score.
segmind ideogram-3
9.1/10 14.9 s
segmind ideogram-3
Cost 0.04 $
Resolution 1344 x 768
Time 14.9 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
7
Fidelity
10
Overall
9.13
Review
The image perfectly respects all elements of the prompt, including the fictional islands, sea monsters, and the compass rose, with a very convincing 17th-century aesthetic. The technical quality is excellent, featuring a parchment texture and very sharp engraving details. The text is stylized consistently with the era, even if its legibility is secondary to the overall artistic rendering.
segmind seedream-4.5
8.1/10 13.2 s
segmind seedream-4.5
Cost 0.04 $
Resolution 2560 x 1472
Time 13.2 s
Matania Judgment
Quality
8
Composition
9
Creativity
8
Text accuracy
4
Fidelity
9
Overall
8.13
Review
The image perfectly respects the requested aesthetics and elements (monsters, compass rose, 17th-century style). The composition is balanced and highly immersive, although the text precision is poor, which is typical of image generation models when dealing with calligraphic inscriptions.
segmind seedream-v5-lite
8.1/10 49.6 s
segmind seedream-v5-lite
Cost 0.04 $
Resolution 2848 x 1600
Time 49.6 s
Matania Judgment
Quality
8
Composition
9
Creativity
8
Text accuracy
4
Fidelity
9
Overall
8.13
Review
The image perfectly respects the requested aesthetics and thematic elements (islands, monsters, compass rose). The composition is balanced and highly evocative of the 17th century, although the text accuracy is poor, which is typical of image generation models. The overall technical quality is excellent, featuring a beautiful parchment texture.
xai grok-imagine-image
8.8/10 6.3 s
xai grok-imagine-image
Cost 0.02 $
Resolution 2752 x 1504
Time 6.3 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
4
Fidelity
10
Overall
8.75
Review
The image is visually stunning and perfectly adheres to all the thematic elements of the prompt (islands, monsters, compass rose, 17th-century style). The technical quality and composition are excellent, although the text accuracy (the lettering typical of old maps) is mediocre, which is common for image generation models.
xai grok-imagine-image-pro
8.8/10 17.0 s
xai grok-imagine-image-pro
Cost 0.07 $
Resolution 2816 x 1536
Time 17.0 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
4
Fidelity
10
Overall
8.75
Review
The image is visually stunning, perfectly adhering to the requested aesthetics and elements (monsters, compass rose, islands). The technical quality and composition are excellent, although text precision is low (illegible or distorted characters), which is typical of image generation models. Prompt fidelity is absolute.

9 emotions portraits

image
google gemini-2.5-flash-image
6.3/10 7.3 s
google gemini-2.5-flash-image
Cost < 0.01 $
Resolution 1344 x 768
Time 7.3 s
Matania Judgment
Quality
8
Composition
7
Creativity
7
Text accuracy
4
Fidelity
6
Overall
6.25
Review
The image respects the 9-portrait grid structure with visual character consistency, but partially fails on textual fidelity as the captions are either illegible or inconsistent with the displayed emotions. While the aesthetic quality is good, the failure to strictly follow the precise captioning instruction heavily impacts the fidelity score.
google gemini-3-pro-image-preview
5.3/10 27.0 s
google gemini-3-pro-image-preview
Cost < 0.01 $
Resolution 2752 x 1536
Time 27.0 s
Matania Judgment
Quality
8
Composition
9
Creativity
7
Text accuracy
2
Fidelity
4
Overall
5.25
Review
While the visual quality and character consistency are excellent, the model fails heavily on the text constraint: the captions are illegible scribbles rather than actual words. Fidelity is penalized because the prompt explicitly required 'captioned' portraits, which is not functionally respected.
google imagen-4.0-fast-generate-001
6.3/10 5.2 s
google imagen-4.0-fast-generate-001
Cost 0.02 $
Resolution 1408 x 768
Time 5.2 s
Matania Judgment
Quality
8
Composition
9
Creativity
7
Text accuracy
2
Fidelity
6
Overall
6.25
Review
The visual quality and character consistency across the portraits are excellent. However, the model fails significantly on textual accuracy, with text being illegible or incoherent, and the grid structure does not perfectly adhere to the request for explicit captions for each emotion.
google imagen-4.0-generate-001
6.4/10 6.1 s
google imagen-4.0-generate-001
Cost 0.04 $
Resolution 1408 x 768
Time 6.1 s
Matania Judgment
Quality
8
Composition
9
Creativity
7
Text accuracy
3
Fidelity
6
Overall
6.38
Review
The image exhibits excellent visual quality and character consistency across the grid. However, fidelity is penalized because the caption text is largely illegible or incoherent, and the grid structure does not always ensure a clear distinction between the 9 requested emotions. Textual accuracy is the major weakness of this generation.
google imagen-4.0-ultra-generate-001
5.9/10 15.0 s
google imagen-4.0-ultra-generate-001
Cost 0.08 $
Resolution 1408 x 768
Time 15.0 s
Matania Judgment
Quality
9
Composition
8
Creativity
7
Text accuracy
3
Fidelity
5
Overall
5.88
Review
The image exhibits excellent visual quality and character consistency, but fails on critical aspects of the prompt. While the portrait grid is respected, the caption text is largely incoherent or illegible, and the exact number of distinct, clearly identified emotions is difficult to verify due to these textual errors.
openai chatgpt-image-latest
6.6/10 43.9 s
openai chatgpt-image-latest
Cost 0.21 $
Resolution 1536 x 1024
Time 43.9 s
Matania Judgment
Quality
9
Composition
9
Creativity
7
Text accuracy
4
Fidelity
6
Overall
6.63
Review
The image is technically excellent, demonstrating high subject consistency and beautiful rendering quality. However, the model fails on textual fidelity and precision: captions are either missing or illegible/incoherent, and the grid does not strictly adhere to the requirement of clearly labeling each emotion.
segmind ideogram-3
9.5/10 15.2 s
segmind ideogram-3
Cost 0.04 $
Resolution 1344 x 768
Time 15.2 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.5
Review
The image perfectly adheres to all the prompt constraints: a 9-portrait grid featuring a consistent person, distinct emotions, and perfectly legible, error-free text. The technical quality is excellent, boasting high sharpness and a well-balanced grid composition.
segmind seedream-4.5
4.9/10 21.1 s
segmind seedream-4.5
Cost 0.04 $
Resolution 2560 x 1472
Time 21.1 s
Matania Judgment
Quality
7
Composition
8
Creativity
6
Text accuracy
2
Fidelity
4
Overall
4.88
Review
The image fails significantly on fidelity and textual accuracy: although it displays a grid of portraits, the captions are completely illegible and consist of incoherent characters (gibberish). Furthermore, the instruction for '9 distinct emotions' is poorly followed, as the expressions are very similar and the text makes it impossible to identify them.
segmind seedream-v5-lite
4.9/10 34.7 s
segmind seedream-v5-lite
Cost 0.04 $
Resolution 2848 x 1600
Time 34.7 s
Matania Judgment
Quality
7
Composition
8
Creativity
6
Text accuracy
2
Fidelity
4
Overall
4.88
Review
The model fails on the major textual constraint, producing illegible or incoherent captions. Although the portrait grid is well-composed and adheres to the concept of the same person, fidelity is heavily penalized by the lack of readable text and the failure to follow the "captioned" instruction.
xai grok-imagine-image
4.9/10 8.1 s
xai grok-imagine-image
Cost 0.02 $
Resolution 2752 x 1504
Time 8.1 s
Matania Judgment
Quality
8
Composition
7
Creativity
6
Text accuracy
2
Fidelity
4
Overall
4.88
Review
The model fails on the most crucial aspect of the prompt: the text. Although the portrait grid is aesthetic and maintains character consistency, the captions are composed of incoherent and illegible characters. Fidelity is heavily penalized because the request for "captions" is not functionally fulfilled.
xai grok-imagine-image-pro
4.9/10 14.9 s
xai grok-imagine-image-pro
Cost 0.07 $
Resolution 2816 x 1536
Time 14.9 s
Matania Judgment
Quality
8
Composition
7
Creativity
6
Text accuracy
2
Fidelity
4
Overall
4.88
Review
The image fails to adhere to textual constraints: the captions consist of incoherent and illegible characters (typical text hallucinations seen in image models). Although the 9-portrait grid is maintained and the subject's identity remains consistent, the lack of textual precision heavily impacts the fidelity and utility of the creation.

Pixel Art

image
google gemini-2.5-flash-image
9.3/10 6.4 s
google gemini-2.5-flash-image
Cost < 0.01 $
Resolution 1344 x 768
Time 6.4 s
Matania Judgment
Quality
9
Composition
8
Creativity
7
Text accuracy
10
Fidelity
10
Overall
9.25
Review
The image perfectly respects the requested 16-bit pixel art style, demonstrating excellent mastery of the color palette and pixel sharpness. The composition is balanced, and the aesthetic is highly consistent with the standards of that era. Despite the minimalist prompt, the model successfully interpreted "elaborate" by creating a detail-rich scene without compromising the style.
google gemini-3-pro-image-preview
9.3/10 41.5 s
google gemini-3-pro-image-preview
Cost < 0.01 $
Resolution 2752 x 1536
Time 41.5 s
Matania Judgment
Quality
9
Composition
8
Creativity
7
Text accuracy
10
Fidelity
10
Overall
9.25
Review
The image perfectly adheres to the technical constraints of 16-bit pixel art, featuring exemplary management of the color palette and pixel sharpness. The composition is balanced, and the aesthetic is highly faithful to the original prompt's intent. As no text is present, the textual accuracy score is at the default maximum.
google imagen-4.0-fast-generate-001
9.3/10 3.6 s
google imagen-4.0-fast-generate-001
Cost 0.02 $
Resolution 1408 x 768
Time 3.6 s
Matania Judgment
Quality
9
Composition
8
Creativity
7
Text accuracy
10
Fidelity
10
Overall
9.25
Review
The image perfectly adheres to the requested 16-bit pixel art style, featuring exemplary color palette management and pixel sharpness. The composition is balanced, and the aesthetic is consistent with the standards of the era depicted. The model perfectly captured the essence of the prompt without adding any superfluous elements.
google imagen-4.0-generate-001
9.3/10 8.7 s
google imagen-4.0-generate-001
Cost 0.04 $
Resolution 1408 x 768
Time 8.7 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 constraints, offering a very consistent and technically clean 16-bit aesthetic. The color management and pixel art precision are excellent, although the scene remains fairly conventional in its creative execution.
google imagen-4.0-ultra-generate-001
9.3/10 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
8
Creativity
7
Text accuracy
10
Fidelity
10
Overall
9.25
Review
The image perfectly adheres to 16-bit pixel art style constraints, featuring a coherent color palette and exemplary sharpness. The composition is well-balanced and the aesthetic is highly faithful to the prompt, although the subject remains somewhat generic.
openai chatgpt-image-latest
9.3/10 54.2 s
openai chatgpt-image-latest
Cost 0.21 $
Resolution 1536 x 1024
Time 54.2 s
Matania Judgment
Quality
9
Composition
8
Creativity
7
Text accuracy
10
Fidelity
10
Overall
9.25
Review
The image perfectly adheres to the requested 16-bit pixel art style, featuring a consistent color palette and era-typical dithering. The composition is well-balanced and the pixel sharpness is excellent, without any modern smoothing artifacts that would break the retro aesthetic.
segmind ideogram-3
9.5/10 15.5 s
segmind ideogram-3
Cost 0.04 $
Resolution 1344 x 768
Time 15.5 s
Matania Judgment
Quality
9
Composition
9
Creativity
8
Text accuracy
10
Fidelity
10
Overall
9.5
Review
The image perfectly adheres to the prompt constraints, delivering a highly convincing and detailed 16-bit aesthetic. The pixel sharpness is exemplary, with no visible artifacts, and the composition is balanced and immersive. The fidelity to the requested pixel art style is absolute.
segmind seedream-4.5
9.3/10 15.7 s
segmind seedream-4.5
Cost 0.04 $
Resolution 2560 x 1472
Time 15.7 s
Matania Judgment
Quality
8
Composition
9
Creativity
7
Text accuracy
10
Fidelity
10
Overall
9.25
Review
The image perfectly adheres to the requested 16-bit pixel art style, featuring a consistent color palette and masterful light management. The composition is well-balanced, and the technical execution avoids the typical artifacts found in diffusion models, delivering a highly successful retro aesthetic.
segmind seedream-v5-lite
8.4/10 27.0 s
segmind seedream-v5-lite
Cost 0.04 $
Resolution 2848 x 1600
Time 27.0 s
Matania Judgment
Quality
8
Composition
7
Creativity
6
Text accuracy
10
Fidelity
9
Overall
8.38
Review
The image perfectly adheres to the requested 16-bit pixel art style, featuring a consistent color palette and a successful retro aesthetic. The technique is clean, although the composition remains quite classic and lacks boldness in terms of originality. Prompt fidelity is excellent, as the model perfectly captured the essence of the request.
xai grok-imagine-image
9.1/10 6.4 s
xai grok-imagine-image
Cost 0.02 $
Resolution 2752 x 1504
Time 6.4 s
Matania Judgment
Quality
8
Composition
8
Creativity
7
Text accuracy
10
Fidelity
10
Overall
9.13
Review
The image perfectly adheres to the requested 16-bit pixel art style, featuring a consistent color palette and exemplary sharpness. The composition is well-balanced, and the technical execution avoids typical compression artifacts, delivering a highly successful retro aesthetic.
xai grok-imagine-image-pro
8.4/10 18.6 s
xai grok-imagine-image-pro
Cost 0.07 $
Resolution 2816 x 1536
Time 18.6 s
Matania Judgment
Quality
8
Composition
7
Creativity
6
Text accuracy
10
Fidelity
9
Overall
8.38
Review
The image perfectly adheres to the requested 16-bit pixel art aesthetic, featuring a consistent color palette and a clean technical render. The composition is pleasing, albeit relatively classic for this style. Prompt fidelity is excellent, as all stylistic elements are present without any superfluous additions.
Code