GPT Image 2 OpenAI Imagen 3.0 Generate 002 Google

Settled by community votes across 6 shared challenges, with an AI judge weighing in on each.

GPT Image 2

27.7 arena score

#4 of 62 in Text-to-Image

Skill signature · Text-to-Image

Imagen 3.0 Generate 002

19.1 arena score

#46 of 62 in Text-to-Image

Vote tally

Where the votes landed

GPT Image 2

0.0%

win rate

Ties

0.0%

Imagen 3.0 Generate 002

100.0%

win rate

0.0% 0.0% ties 100.0%

Shared challenges 6

Challenge by challenge

The strongest take from each model on every shared challenge, with the AI judge's read.

Modern Clean Menu

Text-to-Image

“Modern minimalist restaurant menu design, white background with colorful food photos in grid, sections for appetizers/pizza/mains, bold sans-serif fonts, vibrant accents, clean professional layout for casual dining.”

GPT Image 2

Imagen 3.0 Generate 002

AI Judge Analysis

GPT Image 2

+ Excellent typography with legible, professional menu descriptions.
+ Logical hierarchical layout with clear category sections.
+ High-quality, realistic food photography that fits the restaurant aesthetic.

− The layout is slightly more cramped compared to a true minimalist aesthetic.
− The font style for 'Appetizers' is a bit more decorative than 'bold sans-serif'.

Imagen 3.0 Generate 002

+ Strong adherence to the 'grid' requirement of the prompt.
+ Clean minimalist aesthetic with plenty of white space.
+ Good tonal consistency across food photography images.

− Text consists of garbled, illegible 'Lorem Ipsum' style characters.
− Information hierarchy is confusing with 'Pizza' and 'Mains' labels repeated in multiple spots.
− Food items in the photos don't always correspond logically to the adjacent text categories.

Verdict: GPT Image 2 is the clear winner as it provides a fully functional, professional menu design with legible text and high-quality imagery. While Imagen 3.0 Generate 002 captures a nice grid aesthetic, the garbled text and repetitive, nonsensical layout of the menu sections make it unusable for the requested task.

Magic Burger Explosion: Fiery Photorealism Challenge

Text-to-Image

“Ad for 'Magic Burger'. Dynamic, exploded burger with all components (bun, patty, cheese, lettuce, tomato, sauce) suspended in mid-air. Emphasize photorealistic detail and a sense of motion. Dark, fiery background with glowing embers. Integrate text: 'MAGIC BURGER' as a prominent title, 'LIMITED TIME ONLY' as a secondary message, and '€6.99' in a starburst, all rendered with a fiery, glowing effect.”

GPT Image 2

Imagen 3.0 Generate 002

AI Judge Analysis

GPT Image 2

+ Excellent adherence to the fiery, glowing text effect for all requested messages
+ Superior photorealistic textures on the bun, lettuce, and meat patty
+ Strong sense of motion with sauce droplets and flying embers

− The composition is a bit crowded with the text overlays
− Includes red onions which were not specifically requested but fit the theme

Imagen 3.0 Generate 002

+ Clean layout with the burger as the central focus
+ Accurate exploded view with distinct separation of all layers
+ Good lighting contrast between the burger and the background

− Failed to render 'MAGIC BURGER' with the requested fiery, glowing effect
− Text in the starburst contains gibberish/hallucinated characters
− The meat patty and cheese look slightly more like plastic/CGI than real food

Verdict: GPT Image 2 is the superior choice because it followed all text prompts perfectly, including the specific fiery rendering style for the typography. Imagen 3.0 Generate 002 produced higher-quality ingredient separation, but failed to apply the requested glowing effect to the text and included illegible characters in the price starburst.

Chalkboard Menu

Text-to-Image

“Handwritten-style chalkboard menu in a cozy café, all text rendered in the exact same realistic chalk handwriting style with natural variations in letter size, slight slant, and chalk texture — no printed or digital fonts anywhere on the board. Title at the top in elegant cursive chalk handwriting: ‘TODAY’S SPECIALS – APRIL 30, 2026’. Below it, three menu items also in the same handwritten chalk style: ‘Truffle Mushroom Risotto – $24’, ‘Grilled Octopus with Lemon & Herbs – $28’, ‘Brown Butter Chocolate Chip Cookies – $9’. At the very bottom, smaller text in the identical handwritten chalk style (slightly smaller but still clearly legible with the same handwriting characteristics): ‘All items made fresh daily • Ask about our gluten-free options’. Warm ambient café lighting, visible chalk dust and smudges, realistic handwriting imperfections, no clean printed text anywhere.”

GPT Image 2

Imagen 3.0 Generate 002

AI Judge Analysis

GPT Image 2

+ Excellent adherence to the requested text with zero spelling errors.
+ The chalk texture is highly realistic with authentic variations in stroke pressure and smudging.
+ Perfectly captures the 'cozy café' aesthetic with atmospheric lighting and complementary background elements.

− The slant of the handwriting varies slightly more than a single hand might, but it remains within the prompt requirements.

Imagen 3.0 Generate 002

+ Successfully renders the specific date requested.
+ The blackboard surface has realistic chalk dust residue and streaks.

− Numerous spelling errors including 'Risotttto', 'Octpsuin', and 'Choouip'.
− Repeats menu items unnecessarily, creating a cluttered and confusing layout.
− The header font looks more like a digital outline font than authentic hand-lettered cursive chalk.

Verdict: GPT Image 2 is the clear winner as it followed every instruction perfectly, including complex dish names and specific dates without a single typo. In contrast, Imagen 3.0 Generate 002 struggled significantly with text rendering, producing multiple spelling errors and a repetitive layout that ignored the clean structure requested in the prompt.

The Reversed Rodeo

Text-to-Image

“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”

GPT Image 2

Imagen 3.0 Generate 002

0% wins 0% ties 100% wins

AI Judge Analysis

GPT Image 2

+ Excellent adherence to the 'horse on top' spatial instruction
+ High level of texture detail in the spacesuit and lunar surface
+ Clever use of stirrups and reins attached to the astronaut

− The horse's front legs look slightly awkward growing out of its chest

Imagen 3.0 Generate 002

+ Highly aesthetic and cinematic background lighting
+ Clean rendering of the astronaut and horse

− Completely failed the negative constraint to put the horse on top
− Generic interpretation of a common prompt

Verdict: GPT Image 2 is the clear winner because it correctly followed the specific and difficult spatial instruction to place the horse on top of the astronaut. Imagen 3.0 defaulted to a standard 'astronaut riding a horse' composition, ignoring the core requirement of the challenge.

Isometric Miniature Diorama Scenes

Text-to-Image

“Create a clear, 45° top-down isometric miniature 3D cartoon scene of Japan's signature dish: sushi, with soft refined textures, realistic PBR materials, gentle lighting, on a small raised diorama base with minimal garnish and plate. Solid light blue background. At top-center: 'JAPAN' in large bold text, 'SUSHI' below it, small flag icon. Perfectly centered, ultra-clean, high-clarity, square format.”

GPT Image 2

Imagen 3.0 Generate 002

AI Judge Analysis

GPT Image 2

+ Excellent PBR materials with high-quality textures on the fish and wood
+ Typography and layout are perfectly centered and professional
+ Superior lighting and depth of field that enhances the 3D diorama effect

− Includes some extra stylistic elements like the stone lantern not explicitly requested

Imagen 3.0 Generate 002

+ Successfully captures a soft, stylized 3D cartoon aesthetic
+ Very clean, minimal interpretation of the prompt
+ Accurate text rendering and positioning

− Text is aligned to the top-left rather than top-center as requested
− Visual detail is a bit too simplistic for 'realistic PBR' materials
− Lighting is somewhat flat compared to the other model

Verdict: GPT Image 2 followed the centering and layout instructions perfectly, providing high-quality PBR textures that still maintain a miniature diorama feel. Imagen 3.0 Generate 002 failed the centering instruction for the text and was much more simplistic in its material rendering, though it did capture a charming cartoon aesthetic.

Adorable Baby Animals in Sunny Meadow

Text-to-Image

“Hyper-photorealistic scene of fluffy baby animals—a golden retriever puppy, tabby kitten, baby bunny, and red fox kit—with big expressive eyes and ultra-detailed soft fur, playfully chasing butterflies and tumbling together in a lush wildflower meadow, warm golden sunrise light with god rays and dew sparkles, joyful wholesome vibe, 8K masterpiece.”

GPT Image 2

Imagen 3.0 Generate 002

AI Judge Analysis

GPT Image 2

+ Excellent depiction of motion with the kitten and fox lunging forward.
+ Detailed fur textures that feel soft and realistic.
+ Beautifully rendered backlighting and god rays that enhance the meadow atmosphere.

− The fox's front left paw looks slightly distorted or blending into the grass.

Imagen 3.0 Generate 002

+ Very cute 'tumbling' interaction with the bunny on its back.
+ Clear rendering of the dew sparkles in the grass as requested.
+ The puppy’s face is very well-proportioned and expressive.

− The fox appears somewhat static compared to the 'playfully chasing' prompt.
− Overall composition is a bit more staged and less dynamic than Image A.

Verdict: Both models followed the prompt exceptionally well, capturing all four specific animals and the golden lighting. GPT Image 2 is the preferred choice because it better captures the 'playfully chasing' action and dynamic energy of the scene, whereas Imagen 3.0 Generate 002 feels more like a static portrait of the animals sitting in a field.