OpenAI's legacy image generation model supporting generations, edits with masks (inpainting), and variations
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
DALL-E 2
15.8
arena score
#59 of 62 in Text-to-Image
Skill signature
· Text-to-Image
Imagen 4.0 Generate 001
17.1
arena score
#55 of 62 in Text-to-Image
Vote tally
Where the votes landed
DALL-E 2
0%
win rate
Ties
0%
Imagen 4.0 Generate 001
0%
win rate
Shared challenges
13
Challenge by challenge
The strongest take from each model on every shared challenge, with the AI judge's read.
Geometric Composition
Text-to-Image“A glass cube on a wooden table. Inside the cube is a small blue sphere. On top of the cube sits a red book. A green plant is behind the cube, partially visible through the glass. Soft window light from the left.”
DALL-E 2
Imagen 4.0 Generate 001
AI Judge Analysis
DALL-E 2
- + Clean reflections on the wooden surface.
- + Artistic bokeh effect.
- − Fails to include the red book on top of the cube.
- − The blue sphere is missing, instead making the background object blue.
- − The glass cube contains a red shape rather than holding a blue sphere.
Imagen 4.0 Generate 001
- + Perfect adherence to the spatial relationship of all objects.
- + High photographic realism and texture on the book and table.
- + Accurate rendering of light and reflections within the glass cube.
- − The blue sphere appears to be hovering mid-air inside the cube without support.
Verdict: DALL-E 2 failed significantly on prompt adherence, missing the red book and confusing the color assignments for the sphere and background. Imagen 4.0 Generate 001 followed the prompt perfectly, rendering every object in its correct position with high visual quality.
Candid Street Photography
Text-to-Image“A candid street photo of an elderly Japanese man repairing a red bicycle in light rain, reflections on wet pavement, shallow depth of field, 50mm lens, natural skin texture, imperfect framing, motion blur from passing cars, cinematic but realistic, no stylization.”
DALL-E 2
Imagen 4.0 Generate 001
AI Judge Analysis
DALL-E 2
- + Successfully captures reflections on wet pavement
- + Strong use of shallow depth of field
- − Severely misses the primary subject with excessive blur
- − Anatomical and structural incoherence makes the image unreadable
Imagen 4.0 Generate 001
- + Excellent adherence to all prompt details including age, ethnicity, and action
- + High visual quality with realistic skin textures and water droplets
- + Effective use of background motion blur and bokeh from cars
- − The yellow car has some lighting inconsistencies compared to the rest of the scene
Verdict: DALL-E 2 failed to produce a coherent image, resulting in a blurry mess where the subject is unrecognizable. Imagen 4.0 Generate 001 followed the prompt perfectly, providing a cinematic, high-detail photograph with excellent character realization and atmospheric depth.
Fantasy Warrior
Text-to-Image“Close portrait of a battle-worn paladin in ornate engraved plate armor, hair braided with small beads, faint scars and dirt on the skin, warm torchlight reflecting off metal, shallow depth of field, bokeh sparks, lifelike eyes, highly detailed texture on leather straps and cloth underlayer.”
DALL-E 2
Imagen 4.0 Generate 001
AI Judge Analysis
DALL-E 2
- + Features a very close-up, gritty texture
- − Extremely blurry and low-resolution appearance
- − Fails to render recognizable face, hair, or intricate details
- − Anatomically incoherent and heavily artifacted
Imagen 4.0 Generate 001
- + Excellent adherence to all prompt details including braided hair with beads, scars, and ornate engraving
- + Superior lighting and texture on the leather straps and metal armor
- + High-quality, sharp rendering with clear bokeh sparks
- − The torch flame looks a bit static and disconnected from the character
- − The skin texture is slightly smooth for being 'battle-worn'
Verdict: DALL-E 2 produced a low-quality, abstract image that failed to capture any of the specific character details requested. Imagen 4.0 Generate 001 provided a highly detailed, professional-grade portrait that perfectly executed the complex prompt requirements, from the beaded braids to the intricate engravings in the plate armor.
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.”
DALL-E 2
Imagen 4.0 Generate 001
AI Judge Analysis
DALL-E 2
- + Strong bold sans-serif typography
- + High contrast visual style
- − Fails to follow the grid menu layout requested
- − Food imagery is abstract and unappetizing
- − Text is nonsensical and poorly rendered
Imagen 4.0 Generate 001
- + Perfectly follows requested grid layout with categories
- + High-quality realistic food photography
- + Professional and clean minimalist aesthetic
- − Minor text legibility issues in small print
- − Slightly generic graphic elements
Verdict: Imagen 4.0 Generate 001 delivered a functional and professional menu design with clear sections for Appetizers, Pizza, and Mains as requested, featuring vivid and realistic food photography. In contrast, DALL-E 2 produced an abstract, confusing layout that ignored the specific category requirements and provided unidentifiable food visuals.
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.”
DALL-E 2
Imagen 4.0 Generate 001
AI Judge Analysis
DALL-E 2
- + Successfully captures a sense of intense motion and heat
- + Creative glowing effect on the central subject
- − Text is nonsensical and does not follow the prompt's specific requirements
- − Visual quality is low with significant blur and lack of realistic textures
- − Missing required secondary price and time-limited messaging
Imagen 4.0 Generate 001
- + Excellent adherence to all prompt text requirements with perfect spelling
- + Highly detailed, photorealistic textures especially on the sesame bun and patties
- + Clean composition with professional lighting and background embers
- − The starburst for the price is a bit simplistic in design
Verdict: Imagen 4.0 Generate 001 provides a professional-grade advertisement that perfectly follows all text and content instructions, whereas DALL-E 2 fails to render legible text or distinct ingredients. Imagen 4.0's clarity, ingredient definition, and composition are vastly superior, looking like a finished marketing asset.
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.”
DALL-E 2
Imagen 4.0 Generate 001
AI Judge Analysis
DALL-E 2
- + The chalk texture on the individual strokes looks authentic.
- + The layout feels like a natural, zoomed-in shot of a chalkboard.
- − The text is completely illegible and does not follow the prompt's specific menu requirements.
- − The handwriting style is chaotic and inconsistent rather than elegant.
Imagen 4.0 Generate 001
- + Successfully rendered the specific title and menu items requested in the prompt.
- + The text is clear and legible with a consistent handwriting aesthetic.
- + The wooden frame and overall composition are well-defined.
- − Included meta-commentary from the prompt instructions as text on the board (e.g., 'Elegant chalk hand than into title').
- − The handwriting looks somewhat like a digital font rather than authentic, dusty chalk strokes.
- − Contains spelling errors like 'Berbs' and 'Tittle Menu'.
Verdict: Imagen 4.0 Generate 001 is the clear winner as it successfully follows the prompt's instructions regarding specific text content and dates, whereas DALL-E 2 fails completely to produce legible or relevant words. While Imagen 4.0 mistakenly included some of the prompt's descriptive meta-text on the board, its ability to render the requested menu items makes it far more useful and accurate.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
DALL-E 2
Imagen 4.0 Generate 001
AI Judge Analysis
DALL-E 2
- + Captures a more painterly, surreal atmosphere.
- + Follows the prompt about being set in space with a planetary backdrop.
- − Fails the negative constraint, putting the astronaut on top of the horse.
- − Significantly lower resolution and visual clarity compared to modern models.
- − The horse anatomy is distorted with messy legs and a missing fourth limb.
Imagen 4.0 Generate 001
- + High resolution with crisp, cinematic detail.
- + Intricate textures on the spacesuit and horse's mane.
- + Dynamic composition with vibrant lighting.
- − Fails the negative constraint, putting the astronaut on top of the horse.
- − The horse's hind leg has anatomical issues where it meets the body.
- − Ignored the primary gimmick of the prompt (horse on top).
Verdict: Both DALL-E 2 and Imagen 4.0 failed to follow the specific spatial instruction for the horse to be on top of the astronaut. While both models produced the standard 'astronaut riding a horse' trope, Imagen 4.0 is the clear winner for its superior visual fidelity, cinematic lighting, and detailed textures.
The Capybara Taxi Driver
Text-to-Image“Photorealistic scene inside a yellow New York taxi at night. A capybara is driving, wearing a yellow taxi driver cap and a dark jacket. It has a calm, professional expression and both front paws on the steering wheel. In the back seat sits a human businesswoman in a coat, looking at her phone with a completely normal, bored expression (as if this is just another normal ride). Through the windows you can see the streets of Manhattan at night with blurred lights. Realistic taxi interior, photorealistic, detailed fur and fabric, 35mm lens, night lighting with reflections, shallow depth of field.”
DALL-E 2
Imagen 4.0 Generate 001
AI Judge Analysis
DALL-E 2
- + Attempts the requested layout of a passenger in the back and driver in the front.
- − Severely fails photorealism with significant anatomical distortions in the human face.
- − The capybara is unrecognizable and appears like a low-resolution yellow blur.
- − Heavy artifacts and nonsensical details like the black line across the woman's face.
Imagen 4.0 Generate 001
- + Excellent adherence to all prompt details including the yellow cap, jacket, and paws on the steering wheel.
- + High visual quality with sharp details, realistic lighting, and smooth textures.
- + Perfectly captures the 'bored expression' of the businesswoman sitting in the back seat.
- − The taxi sign is visible through the roof/windshield in a slightly physically impossible manner.
Verdict: Imagen 4.0 Generate 001 followed the prompt perfectly, delivering a high-quality, photorealistic image with correct anatomy and a clear narrative. In contrast, DALL-E 2 produced a distorted and nonsensical image with severe artifacts and failed to render a recognizable capybara or human.
The Halloween Invitation
Text-to-Image“Vintage gothic Halloween party invitation. Dark parchment poster, spooky border with webs and thorns, central glowing jack-o-lantern, bats, twisted trees, moody night sky. Add elegant gothic title text saying "Halloween Party Invitation", a small scroll banner saying "You are invited to a night of frights", and event details at the bottom: Date: 30.10.2026 Time: 7pm Location: The Arches, NYC Spooky but polished, cinematic lighting, square format.”
DALL-E 2
Imagen 4.0 Generate 001
AI Judge Analysis
DALL-E 2
- + Strong vintage parchment texture
- + Authentic decayed gothic aesthetic
- − Text is illegible and contains severe misspellings
- − Missing specific requested details like the date and location
- − Low resolution with messy artifacts
Imagen 4.0 Generate 001
- + Perfect text rendering for all requested details and dates
- + Clear and organized composition following all prompt instructions
- + High visual quality with clean illustration and effective lighting
- − Style is more like a modern digital illustration than a 'vintage parchment'
Verdict: Imagen 4.0 Generate 001 is the clear winner as it successfully rendered all specific text requirements, including the date, time, and location, with perfect legibility. In contrast, DALL-E 2 failed to render readable text and missed the core event details, resulting in a muddy and incoherent image.
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.”
DALL-E 2
Imagen 4.0 Generate 001
AI Judge Analysis
DALL-E 2
- + Captures the isometric orientation and light blue background accurately.
- + Minimalist approach consistent with modern 3D icon styles.
- − Failed to render all requested text, showing 'Sush' instead of 'JAPAN' and 'SUSHI'.
- − Missing the flag icon component entirely.
- − The 3D models are very simplistic and lack the 'realistic PBR' quality requested.
Imagen 4.0 Generate 001
- + Exceptional visual quality with highly realistic textures and PBR materials.
- + Perfectly captures the 'miniature' and 'diorama base' aspects of the prompt.
- + Superior sushi rendering with great detail in the rice and fish.
- − Completely ignored all requested text elements ('JAPAN', 'SUSHI', and flag icon).
- − Unsaturated grey background instead of the requested solid light blue.
Verdict: This is a split decision based on prompt adherence versus aesthetic quality. DALL-E 2 attempted the text and background color but failed on quality and spelling, while Imagen 4.0 Generate 001 produced a stunningly high-quality miniature 3D diorama but failed to include any of the requested typography and ignored the background color instructions. DALL-E 2 is technically closer to the layout and composition, but its poor rendering and spelling errors make it less usable than the beautiful, though partially incomplete, output from Imagen 4.0.
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.”
DALL-E 2
Imagen 4.0 Generate 001
AI Judge Analysis
DALL-E 2
- + Natural lighting and realistic grass textures
- + Good sense of motion and action
- − Major anatomical distortions and artifacts, particularly on the animals in the background
- − Failed to clearly include all four requested animals
- − Low overall image clarity and resolution
Imagen 4.0 Generate 001
- + Includes all four requested animals (dog, cat, rabbit, fox) clearly
- + Beautiful, sharp details including dew drops and fur textures
- + Perfect adherence to lighting requests, including god rays and morning sun
- − Composition feels a bit crowded and stylized rather than purely photorealistic
- − The cat's paws look slightly repetitive in their positioning
Verdict: Imagen 4.0 Generate 001 is the clear winner as it successfully rendered all four requested animals with high clarity and detail, whereas DALL-E 2 suffered from severe AI artifacts and anatomical distortions. Imagen 4.0 also captured the specific atmospheric details like dew and morning light much more effectively than DALL-E 2.
Vintage Cafe Logo
Text-to-Image“Vintage minimalist restaurant logo for "Caffè Florian", retro cloche dome with steam and "Est. 1720" banner, classic typography, warm brown and cream tones, subtle texture on light background, vector emblem style.”
DALL-E 2
Imagen 4.0 Generate 001
AI Judge Analysis
DALL-E 2
- + Successfully captures the requested warm brown and cream tones.
- + Provides a vintage, textured paper feel.
- − Text is complete gibberish and does not follow the naming instructions.
- − The cloche icon is poorly formed and lacks the requested steam detail.
- − Composition is cluttered and lacks the minimalist vector look requested.
Imagen 4.0 Generate 001
- + Perfect text rendering for both the brand name and the 'Est. 1720' banner.
- + Clean, minimalist vector aesthetic that perfectly matches the prompt.
- + Clear, balanced composition with accurate iconography including the cloche and steam.
- − The typography is modern-sans-serif rather than a more ornate 'classic' style.
- − The banner is a simple shape rather than a flowing ribbon.
Verdict: DALL-E 2 failed significantly on both text rendering and basic icon formation, producing an illegible and messy design. Imagen 4.0 followed every instruction perfectly, delivering a clean, professional vector logo with precise spelling and an attractive minimalist aesthetic. Imagen 4.0 is the clear winner as it is a ready-to-use design, whereas DALL-E 2's output is unusable.
Apollo 11: Journey to Tranquility
Text-to-Image“Create a clean, modern vector infographic poster about the Apollo 11 mission. NASA-inspired palette (navy, white, muted red, light gray). Flat-vector style, crisp lines, consistent iconography, subtle gradients only. Steps (stop at landing): 1. Launch (Saturn Vicon) 2. Earth Orbit (Earth + orbit ring icon) 3. Translunar (trajectory arc icon) 4. Lunar Orbit (Moon + orbit ring icon) 5. Descent (lunar module descending icon) 6. Landing (lunar module on the surface icon) Small supporting elements (minimal text): • Crew strip: three silhouette icons with only last names: Armstrong, Aldrin, Collins. • Landing site marker: Moon pin labeled "Tranquility" only. Layout constraints: generous margins, large readable labels, clean background with subtle stars. Vector-only, print-poster look, high resolution.”
DALL-E 2
Imagen 4.0 Generate 001
AI Judge Analysis
DALL-E 2
- + Matches the requested NASA-inspired color palette well.
- − Text consists of gibberish and major spelling errors like 'ALLPOO APPLOO'.
- − The layout is chaotic and fails to present a clear 6-step infographic sequence.
- − Heavy artifacts and poorly defined iconography make it difficult to interpret.
Imagen 4.0 Generate 001
- + Excellent adherence to the chronological 6-step mission request with clear labeling.
- + Perfect text rendering for titles and labels.
- + Clean, modern vector aesthetic that aligns perfectly with the flat-design prompt.
- − The 'Translunar' icon is slightly redundant with the main arc.
- − Some minor inconsistencies in the Saturn V rocket's structural details.
Verdict: Imagen 4.0 significantly outperforms DALL-E 2 by providing a coherent, legible, and professionally designed infographic that follows all specific instructions. While DALL-E 2 produces scrambled text and a muddled composition, Imagen 4.0 delivers crisp iconography and accurate text for the Apollo 11 mission steps.
Next steps
Explore each model
Google's latest Imagen 4.0 text-to-image generation model with significantly better text rendering and overall image quality