DALL-E 3 OpenAI vs GPT Image 1 OpenAI

DALL-E 3

• + High artistic detail with intricate wood grain and interior micro-landscape
• + Beautiful cinematic lighting and depth of field
• + High resolution textures on the book and wooden frame

• − Failed to place the red book on top of the cube, placing it inside instead
• − The 'sphere' is complex rather than a simple 'small blue sphere'
• − The cube is more of a wooden frame than a pure glass cube

GPT Image 1

• + Perfect adherence to spatial instructions (book on top, sphere inside)
• + Accurately represents every object in the correct color and position
• + Realistic lighting and shadows consistent with the window on the left

• − The glass cube has some minor structural clipping with the book
• − The sphere appears to be floating rather than resting on the bottom of the cube

DALL-E 3

• + Excellent reflection on the wet pavement mimicking the subject's pose
• + Effective use of the 'imperfect framing' prompt with foreground objects
• + Strong atmosphere with cinematic lighting and environmental details

• − Anatomical issues with the man's feet appearing elongated and distorted
• − The car in the background lacks the requested motion blur, appearing static
• − Skin texture appears slightly painterly rather than 'natural'

GPT Image 1

• + Outstanding natural skin texture and facial realism
• + Realistic lighting and rain droplets visible on the bicycle seat
• + More authentic 50mm lens feel with a credible shallow depth of field

• − Lacks the requested motion blur from passing cars
• − The bicycle chain and wheel structure are physically nonsensical/merged
• − Framing is very conventional rather than the 'imperfect' requested framing

DALL-E 3

• + Excellent high-contrast lighting and warm torchlight reflections
• + Intricate high-detail engraving on the plate armor
• + Powerful close-up composition with vibrant colors

• − Missed the prompt requirement for braided hair with beads
• − Skin texture looks slightly plasticky or digitally smoothed

GPT Image 1

• + Perfect adherence to specific details like braided hair with small beads
• + Extremely realistic skin texture with lifelike dirt and faint scars
• + Strong sense of 'battle-worn' character through facial expression and grime

• − Armor engravings are more subtle and less 'ornate' than requested
• − Overall image is a bit darker, making it harder to see the detailed leather straps

DALL-E 3

• + Provides a comprehensive four-page layout concept
• + Uses a grid-based approach with varied photography
• + Good use of vibrant color blocking/accents

• − Text is largely unintelligible gibberish
• − The 'grid' is somewhat chaotic and cluttered on some pages

GPT Image 1

• + Clean, readable sans-serif typography with legible pricing
• + High-quality, appetizing food photography
• + Perfectly adheres to the 'modern minimalist' and 'white background' brief

• − Missing the 'mains' section specifically requested in the labels (though food is shown)
• − Only shows a single-page view rather than a full menu design concept

DALL-E 3

• + Excellent dynamic lighting with high-contrast fiery effects
• + Complex exploded view with many varied ingredients
• + Great sense of ground impact and flying debris

• − Multiple spelling errors in the text ('MAGIC BURGR', 'Limiited')
• − The price tag lacks the starburst shape and the specific value of '€6.99' requested

GPT Image 1

• + Perfect text accuracy for all components including the price and secondary message
• + Successfully renders the fiery, glowing text effect as requested
• + Clean and photorealistic burger textures

• − The composition is quite static compared to the 'dynamic' request
• − Background is somewhat plain with only minor embers
• − Missing the sense of motion and 'exploded' energy found in the other model

DALL-E 3

• + Excellent artistic flair with decorative chalk illustrations.
• + Realistic lighting and environmental context of a cozy café.

• − Numerous spelling errors like 'Trufle', 'Occtus', and 'Grililled'.
• − The text layout is cluttered and includes nonsensical numbers and words.

GPT Image 1

• + Perfect text rendering with zero spelling errors for all requested items.
• + The chalk texture is highly realistic, showing convincing grain and pressure variations.
• + Adheres strictly to the handwritten requirement without using digital-looking fonts.

• − The composition is a bit plain and lacks the decorative elements requested by 'elegant cursive'.
• − The bottom edge of the board cuts off the final line of text.

DALL-E 3

• + Excellent cinematic lighting and atmosphere
• + The addition of clouds and the Milky Way creates a beautiful surreal aesthetic
• + Clear and sharp rendering of the astronaut suit

• − Failed the negative constraint; the astronaut is on top of the horse
• − The horse's legs and hooves look somewhat distorted and unnatural

GPT Image 1

• + High resolution and realistic texture on the horse's coat
• + Clean composition with good detail on the astronaut's gear
• + Stronger anatomical accuracy for the horse compared to Model A

• − Failed the negative constraint; the astronaut is on top of the horse
• − The surreal element is less creative, adhering to a very standard 'space horse' trope

DALL-E 3

• + Excellent interior detail on the dashboard and upholstery
• + Great textures on the capybara's fur and clothing
• + Interesting lighting with the interior ceiling lamp

• − Failed to include the human passenger requested in the prompt
• − The perspective is more from the passenger's seat than a full scene including the back

GPT Image 1

• + Successfully followed all instructions including the bored passenger
• + High level of photorealism with shallow depth of field
• + Natural cinematic lighting that fits the nighttime city setting

• − The 'paws' look slightly like human fingers in gloves
• − Interior detail of the car is less defined than the other model

DALL-E 3

• + Ornate and intricate gothic border design
• + Highly stylized 3D-depth and cinematic lighting
• + Creative interpretation of twisted trees and thorns integrated into the frame

• − Text is largely illegible gibberish at the bottom
• − Failed to include the specific prompt text correctly, creating words like 'THIGELSTS'

GPT Image 1

• + Excellent text legibility and accuracy for most of the prompt
• + Clearer layout that serves as a functional invitation
• + Mood and lighting better match the vintage gothic request

• − Merged the 'Time' and 'Location' fields into a single confusing line
• − The border is a bit thin and less prominent than requested

DALL-E 3

• + Excellent 3D toy-like aesthetic with high-gloss finish
• + Very clean isometric perspective and lighting
• + Creative integration of text on the side of the diorama base

• − Failed to place text at 'top-center' as requested
• − Missions the word 'SUSHI' in the text elements
• − The sushi design is repeated/monolithic rather than a varied dish

GPT Image 1

• + Perfect adherence to text placement instructions (top-center, dual lines)
• + Includes the small flag icon exactly where requested
• + Features a more realistic variety of sushi types (nigiri and maki)

• − Texture is a bit more matte/clay-like than the 'refined PBR' look of Model A
• − The lighting is slightly darker and less vibrant than Model A

DALL-E 3

• + Successfully includes all four requested animals: puppy, kitten, bunny, and fox kit.
• + Effective use of god rays and cinematic lighting to create a dreamlike atmosphere.
• + Captures a very expressive and whimsical style that aligns with 'big expressive eyes'.

• − The 'butterflies' have strange mammalian bodies and wings, which looks unnatural and eerie.
• − The style leans heavily toward digital illustration rather than the requested 'hyper-photorealistic' scene.
• − The fox kit has unusual black markings on its paws that look like hearts, which is slightly cartoonish.

GPT Image 1

• + Excellent adherence to the 'hyper-photorealistic' requirement with realistic fur and anatomical structures.
• + Successfully captures the action of 'playfully chasing' and 'tumbling' with dynamic poses.
• + Natural-looking lighting and environment that feels like a real meadow.

• − The bunny is missing its back legs or has them merged into the cat's anatomy in an anatomical error.
• − The kitten's tail is missing or obscured in a way that makes the body look slightly truncated.
• − The butterflies are less prominent and lack the 'ultra-detailed' feel requested compared to the animals.

DALL-E 3

• + Excellent visual appeal with high-quality stipple texture
• + More complex and professional composition
• + Perfectly centered and balanced design elements

• − Failed to include the primary text 'Caffè Florian', substituting it with 'Coffee House'
• − The design is quite busy for the 'minimalist' portion of the prompt

GPT Image 1

• + Successfully included all requested text including 'Caffè Florian'
• + Much better adherence to the 'minimalist' and 'vector emblem' style
• + Correct placement of the 'Est. 1720' text on a banner

• − The black background contradicts the 'light background' request
• − The graphics are somewhat simple compared to Model A
• − The steam effect is a bit thin and less stylized

DALL-E 3

• + Strong artistic aesthetic with a vintage space-age feel.
• + Sophisticated use of the requested color palette.
• + High visual complexity and detailed textures within the moons.

• − Failed to follow the step-by-step instruction, instead creating a repetitive layout.
• − Inaccurate iconography, including depicting Space Shuttles which were not part of Apollo 11.
• − Text is mostly illegible gibberish.

GPT Image 1

• + Excellent adherence to the specific 6-step infographic structure.
• + Correct iconography including the Saturn V, Lunar Module, and Earth/Moon orbits.
• + Clean, modern vector style with legible and mostly accurate text.

• − Composition is a bit cluttered and lacks the flow of a professional poster.
• − Minor spelling error in 'EARLLUNAR'.
• − The flat-vector style is somewhat simplistic compared to Model A.