Imagen 3.0 Generate 002 Google Recraft V4 Recraft AI

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

Imagen 3.0 Generate 002

21.0 arena score

#36 of 62 in Text-to-Image

Skill signature · Text-to-Image

Recraft V4

25.6 arena score

#11 of 62 in Text-to-Image

Vote tally

Where the votes landed

Imagen 3.0 Generate 002

win rate

Ties

Recraft V4

win rate

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.”

Imagen 3.0 Generate 002

Recraft V4

AI Judge Analysis

Imagen 3.0 Generate 002

+ Features a sophisticated, modern grid layout.
+ Includes a wide variety of high-quality, professional food photography.
+ Accurately represents the 'casual dining' atmosphere described in the prompt.

− The text is largely gibberish/lorem ipsum, making it non-functional as a menu.
− The layout is somewhat cluttered with multiple repeated sections for 'Pizza' and 'Mains'.
− Text rendering is poor with letters blending into each other.

Recraft V4

+ Excellent text readability with clear, legible fonts and item descriptions.
+ Extremely clean, minimalist white background consistent with modern menu design.
+ Perfect adherence to all requested categories: Appetizers, Pizza, and Mains.

− The layout is a bit repetitive, following a standard vertical list rather than a dynamic grid.
− Some food items (like the 'Calamari Fritti') look slightly generic/clip-art style compared to Model A.

Verdict: Recraft V4 is the superior choice because it produces a functional menu with readable, logical text and clear pricing, whereas Imagen 3.0 generates illegible text despite a more visually interesting photography grid. Recraft V4 also perfectly categorized the items as requested, making it ready for use in a professional context.

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.”

Imagen 3.0 Generate 002

Recraft V4

AI Judge Analysis

Imagen 3.0 Generate 002

+ Excellent food styling with vibrant, appealing colors
+ Perfect rendering of requested text with glowing effects
+ Stronger adherence to the fiery background and ember effects

− Nonsense filler text above the price tag
− The sauce droplets look slightly cartoonish compared to the bun

Recraft V4

+ Natural, photorealistic lighting on the food components
+ Very dynamic and chaotic 'exploded' composition
+ Accurate rendering of the starburst price tag

− Composition feels a bit unbalanced with a large empty dark space at the top
− The background lacks the level of 'fiery' intensity requested
− White plate at the bottom cuts off the motion

Verdict: Imagen 3.0 provides a more polished and professional advertisement layout with superior text integration and a vibrant color palette that matches the fiery theme. Recraft V4 offers a more realistic and dynamic food explosion, but the overall composition and background are less impactful for a commercial concept.

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.”

Imagen 3.0 Generate 002

Recraft V4

AI Judge Analysis

Imagen 3.0 Generate 002

+ High resolution and clear text rendering on the chalkboard surface.
+ Artistic chalk texture and double-lined title lettering.

− Numerous spelling errors including 'Risotlto', 'Octpsuin', and 'Choouip'.
− Repetitive lines of text at the bottom and lack of the requested elegant cursive for the title.

Recraft V4

+ Perfect adherence to spelling and all three menu items, including finishing the truncated prompt text.
+ Excellent composition with decorative chalk illustrations and a realistic café environment.
+ Very realistic light chalk texture on the board surface.

− The text style leans closer to a digital font-face than a variable natural handwriting.
− Does not use 'elegant cursive' for the title as specifically requested.

Verdict: Recraft V4 is the clear winner for its superior ability to handle complex text instructions and spelling, correctly identifying and completing the truncated prompt for the third menu item. While Imagen 3.0 provides a more authentic 'hand-drawn' feel to individual letters, it fails significantly on spelling and follows the prompt's layout and content instructions less accurately.

The Reversed Rodeo

Text-to-Image

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

Imagen 3.0 Generate 002

Recraft V4

AI Judge Analysis

Imagen 3.0 Generate 002

+ Excellent anatomical rendering of the horse
+ High visual clarity and vibrant cinematic lighting
+ Perfect adherence to the core subject matter

− Simple composition with a flat background nebula
− The stirrups and saddle placement look a bit rigid

Recraft V4

+ Dynamic composition with interesting foreground elements
+ Strong sense of motion and scale with the planet background
+ Detailed texture on the astronaut suit and horse's coat

− The horse's front-right leg has an anatomical glitch at the knee
− Dust and debris look slightly messy around the hooves

Verdict: Imagen 3.0 provides a much cleaner and anatomically correct image with beautiful lighting, whereas Recraft V4 offers a more complex and dynamic scene but suffers from structural issues in the horse's legs. Ultimately, Imagen 3.0 is preferred for its technical polish and clear interpretation of the prompt.

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.”

Imagen 3.0 Generate 002

Recraft V4

AI Judge Analysis

Imagen 3.0 Generate 002

+ Excellent adherence to the '3D cartoon' and 'miniature' style requests.
+ Clean and readable text layout that follows the top-center instruction.
+ Consistent soft lighting and refined, smooth textures.

− The text is aligned to the top-left corner instead of being perfectly centered.
− The rice grains look slightly repetitive and pill-like.

Recraft V4

+ Perfectly centered text and flag icon at the top of the frame.
+ Highly realistic PBR materials, especially on the fish and ice base.
+ Creative interpretation of the diorama base using crushed ice.

− Failed to provide the requested 'cartoon' or 'miniature' aesthetic, opting for realism instead.
− Composition feels slightly cluttered within the plate compared to the minimalist request.

Verdict: Imagen 3.0 followed the stylistic instructions for a 'cartoon scene' much better than Recraft V4, which produced a photorealistic result. However, Recraft V4 was superior in its precise centering of the text elements and flag as requested. Imagen 3.0 is the likely winner for capturing the core miniature aesthetic and overall cleanliness of the prompt.

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.”

Imagen 3.0 Generate 002

Recraft V4

AI Judge Analysis

Imagen 3.0 Generate 002

+ Excellent fur texture rendering and realistic lighting
+ Very clean, 8K-like clarity with minimal artifacts
+ Captures the 'tumbling together' aspect of the prompt well

− The fox kit looks slightly like a hybrid between a dog and a fox
− Less dynamic movement than requested in 'chasing butterflies'

Recraft V4

+ Successfully captures the action of 'chasing' with dynamic poses
+ Stronger 'god rays' and atmospheric dew effects
+ Each animal has distinct, accurate species characteristics

− The kitten has slight anatomical warping around the front paws
− The background feels a bit cluttered with a high number of butterflies
− Slightly less 'photorealistic' than Image A, appearing more like high-end digital art

Verdict: Both models followed the prompt exceptionally well, including all four requested animals. Imagen 3.0 Generate 002 produced a more polish, photorealistic image with superior fur textures, while Recraft V4 better captured the energy of the animals chasing butterflies in a more dynamic composition. Recraft V4 is the winner for its better adherence to the action and more distinct character designs.