GPT Image 1.5 OpenAI vs Qwen Image 2.0 Pro Alibaba

GPT Image 1.5

• + Excellent adherence to the spatial requirements of the prompt.
• + Realistic light refraction and caustic reflections on the table.
• + Very high photographic quality and natural textures.

• − The blue sphere is quite large relative to the cube compared to the word 'small' in the prompt.

Qwen Image 2.0 Pro

• + Successfully includes all requested elements.
• + The blue sphere is appropriately 'small' as requested.

• − The sphere is floating unnaturally in the center of the air without support, which lacks physical realism.
• − The glass cube has some geometric inconsistencies where the panels meet.
• − The blue sphere lacks realistic surface reflections compared to the environment.

GPT Image 1.5

• + Excellent adherence to the 'fiery, glowing effect' on all text elements.
• + Superior sense of motion and 'exploded' dynamics with flying ingredients and embers.
• + Highly detailed food textures, particularly the seared patty and fresh-looking tomatoes.

• − The composition is very crowded, making some text slightly harder to read against the busy background.

Qwen Image 2.0 Pro

• + Clean, professional layout that clearly separates the marketing text from the product.
• + Effective use of a starburst element for the price as requested.
• + Accurate text rendering and placement.

• − The burger feels static rather than 'exploded' with motion as requested.
• − Lower internal detail on the burger components compared to Model A.
• − The starburst and price lack the requested 'fiery, glowing effect', appearing as flat vector graphics.

GPT Image 1.5

• + Excellent chalk texture with natural-looking smudges and dust.
• + Very consistent handwriting style across the whole board.
• + High fidelity to the requested text and date.

• − The handwriting appears slightly more like a digital brush than a physical person holding chalk.
• − The composition is a bit flat with no background context of the 'cozy café'.

Qwen Image 2.0 Pro

• + Stronger sense of place with the cafe background and lighting.
• + The handwriting looks very authentic to a physical person writing with chalk, including natural variations in pressure.
• + Excellent composition with text wrapping that feels realistic for a small board.

• − The lettering is slightly less sharp and crisp compared to Image A.
• − There is a small amount of digital artifacting around some of the thinner lines.

GPT Image 1.5

• + Excellent photorealistic texture on the capybara's fur.
• + Stronger cinematic lighting and depth of field that makes the night scene feel authentic.
• + Perfectly captures the 'bored' expression of the passenger in the background.

• − The capybara's paws look slightly more like humanoid fingers than natural capybara feet.

Qwen Image 2.0 Pro

• + Successfully places both the driver and passenger in a clear, wide composition.
• + Accurately represents the requested yellow taxi driver cap and dark jacket.

• − The passenger appears to be in the front passenger seat rather than the back seat as requested.
• − The capybara's hand on the wheel has a greenish, sickly skin tone and odd anatomy.
• − Overall lighting feels a bit flat and less 'photorealistic' compared to the competition.

GPT Image 1.5

• + Excellent adherence to the 'vintage dark parchment' texture and sepia-toned gothic aesthetic.
• + Flawless text rendering for both the large title and the small event details.
• + Superior composition with a more natural integration of the thorny border and misty graveyard background.

• − The bats are somewhat simplified silhouettes compared to Model B.

Qwen Image 2.0 Pro

• + Features detailed, characterful bats and a unique green internal glow for the pumpkin.
• + Clear typography and includes all requested text elements correctly.

• − The digital art style feels more modern and clean than the requested 'vintage' and 'dark parchment' aesthetic.
• − The scroll banner looks slightly overlaid and less integrated into the environment.

GPT Image 1.5

• + Excellent PBR material rendering with realistic wood, ceramic, and food textures.
• + Detailed and visually rich scene including extra props like a teapot and soy sauce.
• + Strong 3D isometric perspective on a well-defined diorama base.

• − Included significantly more items than the 'minimal garnish' requested.

Qwen Image 2.0 Pro

• + Adhered strictly to the 'minimal garnish' and simple diorama base instructions.
• + Text is clear and the flag icon is correctly placed.
• + Soft, clean cartoonish aesthetic that fits the 'miniature 3D cartoon' prompt.

• − The textures are very basic and lack the 'realistic PBR' quality requested.
• − Composition is a bit sparse and lacks the visual appeal of a miniature scene.
• − Layout of text is slightly awkwardly spaced compared to the graphic design in the other image.

GPT Image 1.5

• + Expertly captures the 'big expressive eyes' and 'tumbling together' aspects of the prompt.
• + Superior lighting with beautiful god rays and sparkling dew effects.
• + High-level rendering of fur textures and paw pad details.

• − The fox kit's anatomy is slightly merged with the other animals in the huddle.
• − The composition is very tightly cropped, losing some of the 'meadow' environment.

Qwen Image 2.0 Pro

• + Excellent full-body composition that shows all four animals clearly in the meadow.
• + Realistic anatomical proportions for the fox and rabbit.
• + Captures the interaction and playful chasing vibe well.

• − The kitten has strange, slightly distorted facial features.
• − The lighting and 'sparkle' effects are more muted and less 'hyper-photorealistic' than the other version.

GPT Image 1.5

• + Excellent typography with correct spelling and accent mark.
• + Professional vintage look with sophisticated shading and texture.
• + Strong composition that follows a classic vector emblem hierarchy.

• − Failed to provide a light background as requested, using black instead.

Qwen Image 2.0 Pro

• + Successfully used a light background with subtle texture.
• + Clean, minimalist vector aesthetic that matches the prompt style.

• − Spelling error in the main text ('Floriian' instead of 'Florian').
• − The cloche handle and steam design are somewhat generic compared to Model A.

GPT Image 1.5

• + Excellent typography and spelling of all stages.
• + Clean, professional panel-based layout that is easy to read.
• + Accurate representation of the astronauts' names and icons.

• − The 'Translunar' icon is slightly confusing as it shows a tiny moon next to a giant lunar orbit icon.
• − The Saturn V rocket in the launch panel is stylized to the point of looking a bit squashed.

Qwen Image 2.0 Pro

• + Elegant vertical flow that better represents the physical journey from Earth to Moon.
• + Includes an actual poster border, making it feel more like a finished piece of graphic design.
• + Accurate and detailed icons for the Lunar Module.

• − Small text under the astronaut icons is blurry and less legible.
• − The S-curve for the 'Translunar' trajectory is a bit abstract compared to a direct arc.
• − Includes a random astronaut face in the icon row that breaks consistency with the silhouettes.