Google's Imagen 3.0 text-to-image generation model, producing high-quality images with improved detail and lighting
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 V4.1
17.6
arena score
#53 of 62 in Text-to-Image
Vote tally
Where the votes landed
Imagen 3.0 Generate 002
0%
win rate
Ties
0%
Recraft V4.1
0%
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 V4.1
AI Judge Analysis
Imagen 3.0 Generate 002
- + Expertly follows the grid layout request
- + High-quality, professional food photography
- + Consistent visual style across all tiles
- − Text is mostly gibberish or illegible
- − Layout feels cramped for a functional menu
- − Redundant sections (multiple 'Pizza' headers)
Recraft V4.1
- + Excellent text rendering with legible, relevant menu items
- + Clean, functional minimalist hierarchy
- + Accurate representation of the requested sections (Appetizers/Pizza/Mains)
- − Food photos are small and not aligned in a strict grid as requested
- − Composition leaves significant empty space in the center
Verdict: Recraft V4.1 produces a far more functional and realistic menu with perfectly legible English text and clear pricing, though it struggles to implement the 'grid' requirement effectively. Imagen 3.0 creates a beautiful visual mosaic that follows the grid prompt well, but the text is completely nonsensical and unusable for a real design project. Recraft V4.1 is the winner for its superior text handling and logical adherence to menu design principles.
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 V4.1
AI Judge Analysis
Imagen 3.0 Generate 002
- + Excellent photorealistic texture on the bun and patty
- + Text integration is clean and professional
- + Composition feels balanced and dynamic with the centered explosion
- − Included irrelevant gibberish text inside the starburst
- − The burger looks more stacked than truly 'exploded' compared to the other model
Recraft V4.1
- + Perfectly executed fiery/glowing effect on the 'MAGIC BURGER' title
- + More chaotic and dynamic 'exploded' motion of the ingredients
- + Accurate rendering of all requested text strings
- − The starburst for the price looks like a flat graphic compared to the 3D scene
- − Image quality on the lettuce and bun is slightly less crisp than Model A
Verdict: Recraft V4.1 followed the complex text and styling prompts more accurately, particularly the fiery effect on the titles and the price starburst. Imagen 3.0 produced a more polished, commercial-grade photorealistic image but failed with garbled text in the starburst and a less energetic 'explosion' effect.
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 V4.1
AI Judge Analysis
Imagen 3.0 Generate 002
- + Excellent chalk-like texture with smudges and erasing marks.
- + Accurate date and prompt-matching currency symbols.
- + Clean composition within a wooden frame.
- − Several spelling errors like 'Risotttto', 'Octpsuin', and 'Choouip'.
- − Repeats menu items with slight variations instead of a secondary list.
- − The title font feels slightly more like a digital outline font than pure chalk.
Recraft V4.1
- + Perfect spelling for all requested menu items and the title.
- + Highly realistic chalk handwriting with natural variations in letter size.
- + Excellent execution of the 'Brown Butter' item continuation requested in the prompt.
- − The frame of the chalkboard is cut off at the edges.
- − Slightly less 'smudge' texture compared to Model A, though still very realistic.
Verdict: Recraft V4.1 is the clear winner as it followed the text instructions perfectly, including the completion of the 'Brown Butter' item, without any spelling errors. In contrast, Imagen 3.0 struggled with spelling (e.g., 'Risotttto', 'ebbs') and repetitive text layout, despite having a slightly more atmospheric chalkboard texture.
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 V4.1
AI Judge Analysis
Imagen 3.0 Generate 002
- + Excellent visual clarity and clean rendering of the space background.
- + Balanced composition with high cinematic appeal.
- + Realistic textures on the horse's coat and mane.
- − Completely failed the semantic instruction to have the horse on top of the astronaut.
Recraft V4.1
- + Strong lighting effects and vibrant use of color in the nebula.
- + Good level of detail in the asteroid environment.
- + Highly dynamic pose for the astronaut and horse.
- − Completely failed the semantic instruction to have the horse on top of the astronaut.
- − Minor artifacting where the horse's tail meets the asteroids.
Verdict: Both Imagen 3.0 and Recraft V4.1 failed the negative constraint and logic reversal in the prompt, which requested the horse to be on top of the astronaut. Because both models produced the standard trope of an astronaut riding a horse, Imagen 3.0 is slightly preferred for its superior image quality, cleaner composition, and lack of visual artifacts.
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 V4.1
AI Judge Analysis
Imagen 3.0 Generate 002
- + Excellent 3D cartoon aesthetic with soft, clay-like textures.
- + Very clean isometric composition and lighting.
- + Includes a good variety of sushi types on the diorama.
- − Text is placed in the top-left rather than top-center as requested.
- − The 'SUSHI' text is larger and more prominent than 'JAPAN', reversing the hierarchy requested.
Recraft V4.1
- + Perfect adherence to text placement and hierarchy in the top-center.
- + Impressive material realism on the fish while maintaining the 3D diorama feel.
- + Accurate 45-degree isometric perspective and shadow.
- − The diorama surface has a slightly distracting texture pattern.
- − The color of the salmon is very similar to the tuna, reducing visual contrast.
Verdict: Both models followed the prompt well, but Recraft V4.1 is the winner due to its superior adherence to the specific layout instructions, including the top-center text placement and font weight hierarchy. While Imagen 3.0 captured a charming cartoon style, it failed to center the text and flag as requested.
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 V4.1
AI Judge Analysis
Imagen 3.0 Generate 002
- + Excellent character rendering with very expressive, soulful eyes.
- + Follows the prompt for four distinct animal types correctly.
- + Vibrant color palette and beautiful dew sparkle effects on the grass.
- − The posing is a bit static and looks more like a posed photo than active 'playful chasing'.
- − Anatomical oddity with the rabbit's paws appearing somewhat jumbled.
Recraft V4.1
- + Dynamic composition that perfectly captures the 'playfully chasing' aspect of the prompt.
- + Superior lighting effects with clear 'god rays' and backlighting on the fur.
- + Excellent fur texture and realistic interaction with the environment.
- − The fox's eyes look slightly glassy and less expressive compared to the others.
- − The kitten's front paw is a bit blurry/poorly defined.
Verdict: Recraft V4.1 is the winner as it better captures the action requested in the prompt, showing the animals actively jumping and reaching for a butterfly, whereas Imagen 3.0 feels like a still portrait. Recraft V4.1 also delivers much stronger atmospheric lighting and 'god rays' which were explicitly requested.
Next steps
Explore each model
Recraft's V4.1 standard tier text-to-image model — refines V4's photorealism with more natural lighting, softer gradients, and sharper illustration styles for everyday creative work