GPT Image 1 OpenAI vs Qwen Image Max Alibaba

GPT Image 1

• + Excellent adherence to lighting instructions with natural shadows and highlights.
• + High clarity and realistic textures on the glass edges and book cover.
• + Clean, logical composition with a well-defined plant in the background.

• − The sphere appears to be floating without a clear point of contact or support inside the cube.

Qwen Image Max

• + Features realistic light refractions and reflections within the glass panels.
• + Includes a complex shadow pattern on the table denoting window light.

• − The sphere is unrealistically levitating in the center of the cube.
• − The book has strange black graphical artifacts/marks on the cover not mentioned in the prompt.

GPT Image 1

• + Exquisite engraving detail on the plate armor with natural-looking wear.
• + The soft lighting and skin texture appear highly photorealistic.
• + Subtle, realistic interpretation of battle-worn features like dirt and faint scars.

• − The braid beads are very simple and less prominent than requested.
• − Leather and cloth details are mostly obscured by the armor design.

Qwen Image Max

• + Excellent adherence to all prompt elements, including prominent beads and leather straps.
• + Strong emotional expression and clear story-telling through the 'battle-worn' character design.
• + Includes the actual torch source to justify the warm lighting and bokeh.

• − Some skin textures and scars look slightly over-sharpened or artificial.
• − The bokeh sparks are a bit uniform and busy, distracting from the face.

GPT Image 1

• + Excellent typography with clean, consistent glowing effects
• + Perfectly centered and organized 'exploded' layout
• + High photorealistic texture on the patty and vegetables

• − Missed the '6' in the price starburst, displaying '.99' instead
• − The background is somewhat static compared to the requested motion

Qwen Image Max

• + Successfully included all requested text and the correct price
• + Dynamic sense of motion with flying embers and smoke effects
• + Vibrant, high-contrast colors that fit the 'fiery' theme

• − The 'exploded' view is less clear, as components overlap heavily
• − The 'starburst' for the price is a bit messy and over-saturated

GPT Image 1

• + Excellent fur texture and lighting integration on the capybara.
• + The capybara's expression and posture perfectly match the 'calm, professional' prompt.
• + High cinematic quality with realistic depth of field and bokeh windows.

• − The paws on the steering wheel look slightly anthropomorphized and thick.
• − The lady in the back is a bit out of focus compared to the rest of the scene.

Qwen Image Max

• + Detailed taxi interior including dashboard, sun visor, and seatbelt.
• + Fuller view of the Manhattan street scene through the front windshield.
• + Capture of the passenger's bored expression is very accurate to the prompt.

• − The capybara's hands look more like human/monkey hands with fur rather than capybara paws.
• − The capybara's head has a strange anatomical transition where it meets the neck.
• − The textures on the interior ceiling look cracked and damaged, which wasn't requested.

GPT Image 1

• + Excellent atmospheric lighting and moody color palette.
• + Perfectly rendered gothic title font.
• + Precise execution of the requested dark parchment texture.

• − Confused the event details, putting the location where the time should be.
• − The background elements like trees and moon are very faint and lack detail.

Qwen Image Max

• + Highly accurate adherence to all text instructions, including date, time, and location.
• + Richly detailed Illustration with prominent thorns and spiderwebs in the border.
• + Clearly defined central glowing jack-o-lantern and twisted trees.

• − The parchment texture is a bit clean, leaning more toward a modern digital illustration than 'vintage'.
• − The scroll banner is slightly warped at the ends.

GPT Image 1

• + Perfectly centered text layout according to the prompt hierarchy.
• + Excellent clay-like 'cartoon' textured finish that looks high-quality.
• + Very clean, minimal garnish and aesthetic balance.

• − The 'JAPAN' text feels a bit flat compared to the 3D scene.
• − The rice grains look like uniform bumps rather than realistic grains.

Qwen Image Max

• + Impressive realistic PBR materials, especially the glass base and fish shine.
• + Diverse selection of sushi types with high detail.
• + Text has a nice 3D pop with drop shadows.

• − The flag icon is placed to the side rather than top-center as requested.
• − The 'JAPAN' text is slightly clipped at the top edge of the frame.
• − The garnish (spring onions) looks a bit cluttered compared to the 'minimal' request.

GPT Image 1

• + Excellent adherence to the list of animals including the rabbit.
• + Very persuasive lighting with soft god rays and realistic bokeh.
• + Anatomy is relatively consistent for all featured animals.

• − The fox has slightly blurred, dark paws that look a bit like blobs.
• − The golden retriever's back leg/tail area is a bit ambiguous in the background.

Qwen Image Max

• + Successfully captures the 'tumbling' and playful interaction between animals.
• + Lush, colorful wildflower meadow with high detail in the flora.
• + Vibrant butterfly colors.

• − Missing the baby bunny entirely.
• − Anatomical issues including two separate golden retriever-like dogs (or one with a detached body) and a fox with an extra-long, awkwardly placed limb.
• − The fox's face looks slightly more like an adult than a kit.

GPT Image 1

• + Clean vector aesthetic with perfect spelling.
• + Maintains high contrast suitable for a real logo emblem.

• − Ignores the request for a light background, opting for black.
• − The steam element is overly simplistic and looks a bit isolated.

Qwen Image Max

• + Perfect adherence to the light background and subtle texture requirements.
• + Excellent classic typography and detailed steam illustration.
• + Accurate color palette using the requested warm brown and cream tones.

• − Slightly less 'minimalist' than Model A due to the shading lines.