Head to head
Esc

Models · slot A

to navigate to pick

Stable Diffusion 3.5 Medium Stability AI Wan 2.5 (Preview) Alibaba

Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.

Stable Diffusion 3.5 Medium

16.8 arena score

#58 of 62 in Text-to-Image

Skill signature · Text-to-Image

Wan 2.5 (Preview)

23.4 arena score

#27 of 62 in Text-to-Image

Vote tally

Where the votes landed

Stable Diffusion 3.5 Medium

0%

win rate

Ties

0%

Wan 2.5 (Preview)

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.”

Stable Diffusion 3.5 Medium
Wan 2.5 (Preview)

AI Judge Analysis

Stable Diffusion 3.5 Medium

  • + Clean glass material rendering
  • + Coherent window lighting
  • Failed to place red book on top of cube
  • The red object and sphere are inside the cube incorrectly
  • Plant is on the side, not behind and visible through the glass

Wan 2.5 (Preview)

  • + Perfect adherence to complex spatial instructions
  • + Excellent texture on the red book and wooden table
  • + Accurate reflections and refractive effects through the glass
  • Some digital noise/dust particles that might look like artifacts

Verdict: Stable Diffusion 3.5 Medium failed the spatial reasoning of the prompt, placing the 'book' (which looks like a red block) inside the cube rather than on top. Wan 2.5 (Preview) followed every detail of the prompt perfectly, including the specific positioning of the sphere, book, and the plant seen through the glass.

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.”

Stable Diffusion 3.5 Medium
Wan 2.5 (Preview)

AI Judge Analysis

Stable Diffusion 3.5 Medium

  • + Captures a strong sense of mood and environment through lighting and rain texture.
  • + Achieves the 'imperfect framing' requested in the prompt with a natural candid feel.
  • + Reflections on the pavement are very realistic and atmospheric.
  • The subject's hands and the bicycle's handlebars are poorly formed and anatomically incorrect.
  • Lacks the specific 'motion blur from passing cars' requested as the cars appear stationary.

Wan 2.5 (Preview)

  • + Excellent anatomical detail in the hands, face, and skin texture.
  • + Strong adherence to the 'repairing' action with visible tools and a realistic pose.
  • + High resolution and clarity with a convincing 50mm shallow depth of field effect.
  • The cars in the background lack motion blur, appearing as static parked or idling vehicles.
  • The lighting is a bit flat compared to the cinematic shadows in Image A.

Verdict: Wan 2.5 (Preview) produced a far superior technical image with realistic skin textures and believable hands, successfully showing the man actually repairing the bike. While Stable Diffusion 3.5 Medium captured a moodier atmosphere and better 'candid' framing, it failed significantly on the anatomical details of the hands and the bicycle's structure.

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.”

Stable Diffusion 3.5 Medium
Wan 2.5 (Preview)

AI Judge Analysis

Stable Diffusion 3.5 Medium

  • + Excellent engraving details on the armor and rich texture on the leather straps.
  • + Captures a very intense, lifelike gaze with sharp clarity.
  • + Strong lighting composition that highlights the facial structure and armor metallic sheen.
  • Failed to include the beads in the braided hair as requested.
  • The 'bokeh sparks' look more like static dots than atmospheric particles.

Wan 2.5 (Preview)

  • + Accurately included the small beads within the braided hair.
  • + Superior atmospheric lighting with a visible torch source and natural bokeh.
  • + More convincing 'battle-worn' appearance with heavier smearing of dirt and more rags on the underlayer.
  • The facial features are slightly smoother and less detailed compared to Model A.
  • The armor engraving, while good, is slightly less intricate than its counterpart.

Verdict: Wan 2.5 (Preview) followed the prompt more meticulously by including the hair beads and providing a better sense of atmospheric torchlight. Stable Diffusion 3.5 Medium produced a sharper face and more intricate engraving, but its omission of specific prompt details like the beads makes Wan 2.5 (Preview) the better interpretation overall.

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.”

Stable Diffusion 3.5 Medium
Wan 2.5 (Preview)

AI Judge Analysis

Stable Diffusion 3.5 Medium

  • + Clean grid layout for food photography
  • + Consistent use of professional sans-serif fonts
  • + Good variety of dish types shown
  • Text is mostly gibberish or heavily distorted
  • Poor contrast between some text elements and the white background
  • Cluttered bottom section with unintuitive pricing columns

Wan 2.5 (Preview)

  • + Excellent adherence to category labels (Appetizers, Pizza, Mains)
  • + High quality food photography with consistent lighting
  • + Vibrant color coding for different menu sections
  • Some spelling errors in header text
  • The 'Mains' section repeated a pizza slice which feels redundant
  • Food styling is slightly repetitive with heavy use of basil leaves

Verdict: Wan 2.5 (Preview) follows the prompt more accurately by clearly organizing the menu into the requested categories (Appetizers, Pizza, Mains) with bold, legible headers. Stable Diffusion 3.5 Medium provides a realistic layout but fails at text legibility and logical organization, whereas Wan 2.5 creates a functional, aesthetically pleasing design suitable for a casual dining environment.

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.”

Stable Diffusion 3.5 Medium
Wan 2.5 (Preview)

AI Judge Analysis

Stable Diffusion 3.5 Medium

  • + Very realistic textures on the bun and patty.
  • + Excellent rendering of fire and glowing embers in the background.
  • + Correct text spelling and legible starburst price tag.
  • Failed the 'exploded burger' instruction as the ingredients are still stacked together.
  • Text rendering is basic and lacks the requested fiery, glowing effect.

Wan 2.5 (Preview)

  • + Perfectly captured the 'exploded' view with components flying in mid-air.
  • + Executed the 'fiery, glowing effect' for all text elements beautifully.
  • + Strong adherence to the starburst container and dark, smoky atmosphere.
  • The 'LIMITED TIME ONLY' text is slightly warped along the curve.
  • Minor lighting inconsistency on the bottom bun.

Verdict: Stable Diffusion 3.5 Medium produced a higher quality, photorealistic image of a burger, but it failed to follow the primary 'exploded' composition instruction. Wan 2.5 (Preview) followed every single prompt detail, including the specific exploded layout and the fiery stylized text effects, resulting in a much more effective advertisement.

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.”

Stable Diffusion 3.5 Medium
Wan 2.5 (Preview)

AI Judge Analysis

Stable Diffusion 3.5 Medium

  • + Excellent chalk texture and realistic smudging effects.
  • Text is largely nonsensical and illegible.
  • Failed to follow the specific text instructions and date.
  • Layout is cluttered and lacks the requested elegance.

Wan 2.5 (Preview)

  • + Perfect text rendering with near-flawless spelling and adherence to the prompt.
  • + Great sense of depth and a cozy cafe background.
  • + Handwriting style is consistent and realistic while remaining legible.
  • Missed the 'elegant cursive' requirement for the title, opting for all-caps sans-serif instead.
  • Added its own footer text not requested in the prompt.

Verdict: Wan 2.5 (Preview) significantly outperformed Stable Diffusion 3.5 Medium by correctly rendering the requested food items and date with high legibility. While Stable Diffusion 3.5 Medium captured the physical texture of chalk more realistically, the failure to produce readable English text makes it unusable for this prompt compared to the coherent and well-composed output from Wan 2.5.

The Reversed Rodeo

Text-to-Image

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

Stable Diffusion 3.5 Medium
Wan 2.5 (Preview)

AI Judge Analysis

Stable Diffusion 3.5 Medium

  • + Successfully followed the logic of a horse on top of an astronaut.
  • + Simple, clean composition with a striking silhouette against the Earth.
  • + Good lighting coherence between the subjects and the background.
  • The horse's legs are anatomically distorted with multiple joints.
  • Lower level of textural detail compared to the competitor.
  • The placement of the rider's legs behind the horse is confusing.

Wan 2.5 (Preview)

  • + High level of cinematic detail in the clouds, galaxy, and horse's mane.
  • + Dynamic pose with a sense of motion and action.
  • + Strong textures on both the space suit and the horse's coat.
  • Failed the negative constraint/spatial instruction: the astronaut is on top of the horse.
  • Anatomical issues with the horse's hind legs merging into the dust.
  • The composition feels slightly cluttered with too many celestial elements.

Verdict: Stable Diffusion 3.5 Medium was the only model to follow the specific spatial instruction of placing the horse on top of the astronaut, though it struggled with the horse's anatomy. Wan 2.5 (Preview) produced a much more visually impressive and detailed image, but completely ignored the prompt's core 'surreal' inversion. Because Stable Diffusion 3.5 Medium actually attempted the difficult prompt logic, it is the more successful response to the challenge.

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.”

Stable Diffusion 3.5 Medium
Wan 2.5 (Preview)

AI Judge Analysis

Stable Diffusion 3.5 Medium

  • + Excellent photographic quality and lighting on the capybara's fur.
  • + Consistent interior taxi color palette.
  • + Good interpretation of the 'jacket' requirement.
  • Missed the prompt requirement for the passenger to be looking at her phone.
  • The steering wheel is barely visible or implied behind the dashboard.

Wan 2.5 (Preview)

  • + Followed all prompt instructions including the phone and the 'bored expression' of the passenger.
  • + Clear and logical positioning of the front paws on the steering wheel.
  • + The background cityscape is more recognizable as a vibrant Manhattan environment.
  • The capybara's paws have somewhat uncanny, monster-like nails.
  • The interior seating layout looks slightly inconsistent with a standard cab.

Verdict: Stable Diffusion 3.5 Medium produces a more aesthetically pleasing image with better fur texture, but it fails to include the requested interaction (the passenger looking at her phone). Wan 2.5 (Preview) successfully follows all prompt elements including the bored businesswoman and the phone, though the hand details on the capybara are slightly less realistic.

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.”

Stable Diffusion 3.5 Medium
Wan 2.5 (Preview)

AI Judge Analysis

Stable Diffusion 3.5 Medium

  • + Features a distinct vintage parchment texture
  • + Includes the twisted trees and bats as requested
  • Numerous spelling errors including 'Halloweeen', 'Inviloween', and 'Loccation'
  • Composition is cluttered with multiple pumpkins instead of a central one
  • Text layout is messy with dots and extra lines

Wan 2.5 (Preview)

  • + Excellent typography with perfect spelling in all requested text
  • + High visual quality with cinematic lighting and a clear central jack-o-lantern
  • + Strict adherence to the banner and border elements like thorns and webs
  • The parchment is more of a background frame than the main surface
  • Trees are pushed to the side, though still present

Verdict: Wan 2.5 (Preview) significantly outperformed Stable Diffusion 3.5 Medium by correctly spelling every piece of text and delivering a polished, professional composition. While Stable Diffusion 3.5 Medium captured the vintage parchment aesthetic well, its severe typos and disjointed layout made the invitation unusable.

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.”

Stable Diffusion 3.5 Medium
Wan 2.5 (Preview)

AI Judge Analysis

Stable Diffusion 3.5 Medium

  • + Features realistic sub-surface scattering and textures on the roe.
  • + Good clarity on the sushi rolls themselves.
  • Missed the flag icon entirely.
  • The text 'SUSHI' is visually distorted and poorly rendered.
  • Lacks the requested 'raised diorama' base.

Wan 2.5 (Preview)

  • + Excellent typography that perfectly matches the prompt requirements.
  • + Accurately includes all elements including the flag and the raised diorama base.
  • + Achieves a high-quality 3D cartoon/stylized aesthetic with clean lighting.
  • The leaf garnish is slightly simplified relative to the rest of the scene.

Verdict: Wan 2.5 (Preview) is the clear winner as it followed every instruction in the prompt, including the specific text placement, the flag icon, and the architectural diorama base. Stable Diffusion 3.5 Medium failed to render the text correctly, missed the flag, and ignored the diorama requirement, providing a flat background instead.

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.”

Stable Diffusion 3.5 Medium
Wan 2.5 (Preview)

AI Judge Analysis

Stable Diffusion 3.5 Medium

  • + Vibrant saturated colors that enhance the 'wholesome' vibe
  • + Beautiful soft bokeh and lighting in the background
  • Failed to include the baby bunny entirely
  • The animals are sitting still rather than 'playfully chasing' or 'tumbling'
  • The kitten looks strangely hybridized with a rabbit, having long pointed ears

Wan 2.5 (Preview)

  • + Excellent prompt adherence including all four requested animals
  • + Dynamic composition that captures the 'chasing' and 'tumbling' action
  • + Successfully rendered god rays and floating dew sparkles as requested
  • The fox's eyes appear unnaturally blue and slightly distorted
  • The lighting on the kitten is a bit flat compared to the other animals

Verdict: Wan 2.5 (Preview) is the clear winner as it followed the complex prompt precisely, including the dog, kitten, bunny, and fox in a dynamic action pose. Stable Diffusion 3.5 Medium failed to include the bunny and produced a static, posed image that didn't capture the requested movement or 'god rays' as effectively.

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.”

Stable Diffusion 3.5 Medium
Wan 2.5 (Preview)

AI Judge Analysis

Stable Diffusion 3.5 Medium

  • + Excellent vintage illustrative style
  • + Intricate banner and linework
  • Text contains spelling errors ('Florrian', 'Caffee', 'Est 170')
  • Fails to clearly represent a cloche dome
  • Included elements like a tea cup not requested

Wan 2.5 (Preview)

  • + Perfect adherence to text requested ('Caffè Florian', 'Est. 1720')
  • + Clear interpretation of the retro cloche dome and steam
  • + Professional, minimalist vector emblem composition
  • Simple, standard design compared to the more artistic background texture

Verdict: Wan 2.5 (Preview) produced a superior logo by accurately rendering all requested text and specifically depicting the cloche dome as requested. Stable Diffusion 3.5 Medium failed significantly on the typography, misspelling the brand name and the date, and its interpretation of the cloche dome was messy and unrecognizable.

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.”

Stable Diffusion 3.5 Medium
Wan 2.5 (Preview)

AI Judge Analysis

Stable Diffusion 3.5 Medium

  • + Successfully uses the requested NASA-inspired color palette.
  • + Good use of space with a large arc representing the trajectory.
  • Text is largely gibberish and does not follow the requested step-by-step labels.
  • Icons do not match the specific requested objects for each stage.
  • The vertical flow is confusing and logically inconsistent for an infographic.

Wan 2.5 (Preview)

  • + Excellent adherence to the requested steps and labels with legible text.
  • + High-quality vector style with clean lines and consistent iconography.
  • + Strong composition that logically follows the mission trajectory.
  • One minor spelling error on 'Tranquility' (missing an 'l').
  • Includes additional elements like astronaut portraits not specifically requested, though they fit the theme.

Verdict: Wan 2.5 (Preview) significantly outperformed Stable Diffusion 3.5 Medium by correctly following all six logical steps of the infographic and rendering legible, accurate text. While Stable Diffusion 3.5 Medium captured the aesthetic mood, its content was nonsensical and failed to display the specific icons requested, whereas Wan 2.5 (Preview) created a professional-grade poster that accurately depicted the Saturn V, Earth, Moon, and Lunar Module.

Next steps

Explore each model