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
#59 of 62 in Text-to-Image
Wan 2.5 (Preview)
#28 of 62 in Text-to-Image
Where the votes landed
DALL-E 2
0%
win rate
Ties
0%
Wan 2.5 (Preview)
0%
win rate
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.”
AI Judge Analysis
DALL-E 2
- + Clean geometric lines on the cube
- + Realistic wood surface reflection
- − Failed to include the red book on top of the cube
- − Replaced the small blue sphere with a massive blue planter
- − Composition is unbalanced and ignores several prompt instructions
Wan 2.5 (Preview)
- + Perfect adherence to all spatial instructions in the prompt
- + High visual quality with realistic lighting and dust particles
- + Accurate rendering of glass transparency and reflections
- − The red book's reflection on the top facet of the cube looks slightly like a blue sphere
Verdict: Wan 2.5 (Preview) followed every detail of the prompt, correctly placing the blue sphere inside the cube and the red book on top, while maintaining high photographic quality. DALL-E 2 failed significantly on prompt adherence, confusing the blue sphere with a large background object and missing the red book entirely.
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.”
AI Judge Analysis
DALL-E 2
- + Successfully captures reflections on wet pavement
- + Embraces 'imperfect framing' with an obscured foreground element
- − Extreme blur makes the subject unrecognizable and the image nearly incoherent
- − Fails to show a face or clear 'elderly Japanese man' features
Wan 2.5 (Preview)
- + Excellent adherence to all prompt details including age, ethnicity, and activity
- + High visual quality with realistic skin textures and rain effects
- + Perfect execution of shallow depth of field and automotive bokeh
- − The composition is a bit too balanced and 'clean' for an 'imperfect framing' request
- − The bicycle kickstand and tool logic is slightly warped upon close inspection
Verdict: Wan 2.5 (Preview) produced a high-quality, cinematic, and realistic image that followed every aspect of the prompt, including the specific cultural and age descriptors. DALL-E 2 interpreted 'shallow depth of field' and 'imperfect framing' so aggressively that the subject is entirely out of focus and the image is unusable.
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.”
AI Judge Analysis
DALL-E 2
- + Features a very close-up, gritty perspective
- + Includes strong bokeh effects
- − Extremely low resolution and lacks clarity
- − Fails to render recognizable lifelike eyes, braids, or facial features
- − Looks more like a photograph of a decaying miniature than a person
Wan 2.5 (Preview)
- + Perfectly follows every instruction including braided hair with beads and engraved armor
- + Exceptional resolution and texture on leather and frayed cloth
- + Lifelike eyes and convincing lighting from the torch source
- − The character looks slightly young for a 'battle-worn' paladin
- − The sparks look a bit like floating orange dots rather than embers
Verdict: Wan 2.5 (Preview) produced a stunningly accurate and high-quality image that captured every detail of the prompt, from the intricate engravings to the specific hair accessories. DALL-E 2 produced an abstract, low-quality image that failed to represent a human face or the requested details clearly.
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.”
AI Judge Analysis
DALL-E 2
- + Strong bold typography creates a modern aesthetic.
- + High-contrast visual design that feels artistic.
- − Fails to follow the layout request, appearing more like a magazine than a menu.
- − Text is completely illegible and nonsensical.
- − Food photos are distorted and unrecognizable within the geometric grid.
Wan 2.5 (Preview)
- + Excellent adherence to all prompt instructions including specific category sections.
- + Clear, high-quality food photography arranged in a logical grid.
- + Professional layout that functions as an actual restaurant menu.
- − Some minor typos in headers and subtext.
- − The inclusion of the literal prompt words 'Modern minimalist' in the title is a slight prompt leakage.
Verdict: Wan 2.5 (Preview) significantly outperforms DALL-E 2 by providing a functional, well-organized menu layout that Includes the requested sections for Appetizers, Pizza, and Mains. While DALL-E 2 produces a more abstract, high-fashion aesthetic, it fails to deliver on the fundamental requirements of a restaurant menu design, whereas Wan 2.5 creates a professional and appetizing result.
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.”
AI Judge Analysis
DALL-E 2
- + Captures the glowing, fiery energy requested in the prompt.
- − Text is largely misspelled and illegible.
- − Visual quality is low with blurry textures and messy composition.
- − Missing specific required elements like the starburst and specific price text.
Wan 2.5 (Preview)
- + Excellent adherence to all prompt elements, including complex text and price starburst.
- + High photorealistic detail in the food textures and splash effects.
- + Professional composition with clean, glowing typography.
- − The 'fiery background' is slightly more subtle than requested, focusing more on embers and smoke.
Verdict: Wan 2.5 (Preview) produced a professional-grade advertisement that followed every instruction, including accurate rendering of multiple text lines and the specific starburst element. In contrast, DALL-E 2 struggled significantly with text legibility and image clarity, resulting in an abstract and messy layout.
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.”
AI Judge Analysis
DALL-E 2
- + Captured a rough chalk-like texture.
- − The text is completely illegible gibberish.
- − The prompt's specific date and menu items were ignored.
- − The image quality is low with significant compression artifacts.
Wan 2.5 (Preview)
- + Exceptional text rendering that perfectly matches the requested date and menu items.
- + Highly realistic chalk aesthetic with smudges and natural variations.
- + Excellent composition and depth of field creating a cozy café atmosphere.
- − Included a price for the third item that was cut off in the prompt, though logically consistent.
Verdict: Wan 2.5 (Preview) significantly outperformed DALL-E 2 by accurately rendering every specific word, date, and price requested in the prompt with high-fidelity legibility. DALL-E 2 failed to produce legible text and ignored the prompt's specific menu content entirely, resulting in a series of nonsensical chalk strokes.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
AI Judge Analysis
DALL-E 2
- + Successfully followed the difficult positional instruction of putting the horse on top of the astronaut.
- − Resolution is very low and contains significant artifacts
- − The composition and artistic style are dated and lacks cinematic polish
- − The horse's anatomical detail is poor and looks like a statue or clay model
Wan 2.5 (Preview)
- + Excellent visual quality with sharp details and vibrant colors
- + Highly cinematic lighting and composition
- + Excellent rendering of textures like the astronaut's suit and the horse's fur
- − Completely failed the negative constraint/specific positional instruction
Verdict: This is a classic trade-off between prompt engineering and visual fidelity. DALL-E 2 successfully misinterpreted the common phrase to place the horse on top of the astronaut as requested, but the image quality is extremely poor. Wan 2.5 (Preview) produced a stunning, high-quality cinematic image, but it fell into the trap of generating the standard 'astronaut riding horse' trope, ignoring the specific positional instruction.
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.”
AI Judge Analysis
DALL-E 2
- + Attempts to show the device and hands in the foreground.
- − Extremely poor visual quality with severe anatomical glitches.
- − The animal does not resemble a capybara and looks more like an abstract statue.
- − Failed to follow the composition instructions, putting the human in the front/side and missing the businessman archetype.
Wan 2.5 (Preview)
- + Excellent adherence to all prompt details, including the cap, jacket, and paws on the steering wheel.
- + High photorealism with convincing fur texture and atmospheric lighting.
- + Perfectly captures the 'bored' expression of the passenger in the back seat.
- − The paws look slightly more like humanoid hands with claws than natural capybara feet.
Verdict: Wan 2.5 (Preview) is the clear winner, delivering a high-quality, photorealistic image that perfectly follows the complex prompt instructions. In contrast, DALL-E 2 produced a low-resolution image with significant anatomical distortions and failed to represent a capybara or a logical car interior.
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.”
AI Judge Analysis
DALL-E 2
- + Successfully captures a weathered, vintage woodcut aesthetic
- + Follows the square format requirement
- − Text is largely nonsensical and garbled
- − Fails to include specific requested elements like the jack-o-lantern and digital clarity
- − Low resolution and messy visual artifacts
Wan 2.5 (Preview)
- + Excellent text rendering for all requested details and dates
- + High visual clarity with sharp details on the pumpkin, thorns, and webs
- + Follows all prompt instructions including the specific banner and border elements
- − The layout is very clean, which may feel a bit less 'vintage' and more like a modern digital graphic
Verdict: Wan 2.5 (Preview) is the clear winner as it perfectly follows all instructions, specifically the complex task of rendering accurate text for the title, banner, and event details. DALL-E 2 fails to produce legible text or the central jack-o-lantern, resulting in a muddy, indistinct 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.”
AI Judge Analysis
DALL-E 2
- + Successfully captures the 45-degree isometric perspective.
- + Maintains a clean, minimalist blue background as requested.
- − Failed to render the text correctly, displaying only 'Sush'.
- − Missing the 'JAPAN' text and the required flag icon.
- − The 3D models for the food items are unrecognizable and low quality.
Wan 2.5 (Preview)
- + Followed all text instructions perfectly, including 'JAPAN', 'SUSHI', and the flag icon.
- + Excellent rendering of 3D miniature textures and soft lighting.
- + Created a high-quality, appealing diorama base and realistic cartoon sushi.
- − The perspective is more eye-level/front-facing than a true 45-degree top-down isometric view.
Verdict: Wan 2.5 (Preview) significantly outperformed DALL-E 2 by adhering to all complex text requirements and delivering a much higher visual quality. While DALL-E 2 captured the isometric perspective better, its failure to render legible text or recognizable sushi makes it much less successful overall.
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.”
AI Judge Analysis
DALL-E 2
- + Features most of the requested animals in a simple layout.
- − Low visual fidelity with severe artifacts throughout the image.
- − Poor anatomy, particularly on the cat and the distorted central butterfly.
- − Missing the red fox kit and clear 'god rays' lighting.
Wan 2.5 (Preview)
- + Excellent adherence to the prompt, including all four specific animals and environmental details.
- + Superior visual quality with ultra-detailed fur and clear, sharp resolution.
- + Beautiful artistic composition with effective use of god rays, dew sparkles, and golden hour lighting.
- − The fox's eyes appear slightly over-processed or unnatural.
- − One butterfly leg is noticeably missing on the rightmost insect.
Verdict: Wan 2.5 (Preview) significantly outperforms DALL-E 2 in every category, delivering a high-definition, coherent scene that perfectly matches the complex prompt requirements. DALL-E 2 struggles with basic anatomy and lacks the 'hyper-photorealistic' quality requested, resulting in a blurry and distorted image.
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.”
AI Judge Analysis
DALL-E 2
- + Matches the simple vector emblem style requested.
- − Text is completely illegible and gibberish.
- − The steam element is poorly rendered and lacks clarity.
- − Low level of detail and polish compared to modern standards.
Wan 2.5 (Preview)
- + Excellent typography with perfect spelling of 'Caffè Florian'.
- + High visual quality with clean vector lines and appropriate texture.
- + Successful inclusion of all prompt elements including the banner, steam, and date.
- − The layout is slightly more complex than 'minimalist' but remains professional.
Verdict: Wan 2.5 (Preview) significantly outperforms DALL-E 2 by correctly rendering the requested text and providing a polished, professional logo design. DALL-E 2 failed on prompt adherence for both the specific text and the date, producing illegible characters and a disjointed composition.
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.”
AI Judge Analysis
DALL-E 2
- + Adheres to the color palette requested.
- + Captures a technical schematic aesthetic.
- − Text is nonsensical and contains severe spelling errors like 'ALLPOO'.
- − The logical flow of the infographic is chaotic and does not follow the requested steps.
- − Visual artifacts and messy line work make it look low-quality.
Wan 2.5 (Preview)
- + Excellent adherence to the sequential steps of the mission.
- + Legible, correctly spelled labels and names.
- + Clean, modern flat-vector aesthetic that perfectly matches the NASA-inspired theme.
- − Includes a second moon graphic which is slightly redundant in the layout.
- − Iconography for 'Descent' and 'Landing' is clustered rather than separated as distinct steps.
Verdict: Wan 2.5 (Preview) produced a professional, legible, and organized infographic that followed every step of the prompt with high-quality vector art. DALL-E 2 failed significantly, producing garbled text and a disorganized layout that lacked any clear progression or utility.
Explore each model
Alibaba's text-to-image and image-to-image generation model from the Wan AI suite, offering high-quality visual generation capabilities