Head to head
Esc

Models · slot A

to navigate to pick

Qwen Image 2512 Alibaba Wan 2.5 (Preview) Alibaba

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

Qwen Image 2512

22.9 arena score

#30 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

Qwen Image 2512

0.0%

win rate

Ties

0.0%

Wan 2.5 (Preview)

100.0%

win rate

0.0% 0.0% ties 100.0%
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.”

Qwen Image 2512
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2512

  • + Excellent depiction of the glass cube structure
  • + Correct placement of all objects according to the prompt
  • + Highly realistic textures on the book and table
  • The glass has a heavy cyan/green tint rather than clear transparency

Wan 2.5 (Preview)

  • + Beautiful lighting and atmosphere with visible dust motes
  • + Clearer glass transparency showing the plant behind more naturally
  • + Great composition and depth of field
  • The blue sphere is physically impossible, as it is floating mid-air inside the cube
  • The lighting direction is a bit diffuse compared to the requested 'from the left' window light

Verdict: Both models followed the prompt instructions perfectly regarding object placement. Qwen Image 2512 produces a more physically grounded scene where the sphere sits on the floor of the cube, whereas Wan 2.5 (Preview) depicts a floating sphere. However, Wan 2.5 has superior lighting and much more realistic clear glass compared to the tinted glass in Qwen.

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.”

Qwen Image 2512
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2512

  • + Excellent application of motion blur on background cars
  • + Very realistic skin texture and facial details
  • + Strong cinematic atmosphere with natural-looking rain and lighting
  • The subject is looking directly at the camera, which feels less like a 'candid' shot
  • The bicycle structure has some minor clipping issues near the pedals

Wan 2.5 (Preview)

  • + Natural candid pose that fits the 'repairing' instruction perfectly
  • + Beautiful wet pavement reflections and clear tools on the ground
  • + Great inclusion of the bike kickstand and surrounding street context
  • Lack of motion blur on the background cars despite the specific prompt request
  • Raindrops look more like static streaks compared to Model A

Verdict: Qwen Image 2512 followed the technical photography prompts much better, particularly the motion blur and the gritty, photorealistic texture. However, Wan 2.5 captured a more authentic candid moment of the man actually performing a repair, whereas Qwen Image 2512 looks like a posed portrait.

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.”

Qwen Image 2512
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2512

  • + Excellent depiction of a battle-worn facial texture with realistic scars and dirt.
  • + Intricate engraving on the plate armor that looks heavy and authentic.
  • + Superior lighting effects with high-quality bokeh sparks and warm torchlight reflection.
  • The braids are a bit messy and less defined compared to Model B.

Wan 2.5 (Preview)

  • + Clear and distinct braided hair with well-defined bead elements.
  • + Strong contrast in textures between the metal, rough cloth, and leather straps.
  • The character looks significantly younger and less experienced than the requested 'battle-worn' trope.
  • Light bokeh in the background looks like simple orange circles rather than natural sparks.

Verdict: Qwen Image 2512 captures the 'battle-worn' essence much better than Wan 2.5 (Preview), providing a face that looks like it has seen many combats. Qwen also provides more realistic environmental effects like the sparks and the way torchlight hits the metal, whereas Wan 2.5 (Preview) looks a bit more like a clean studio portrait of a young actor.

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.”

Qwen Image 2512
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2512

  • + Strong professional layout resembling a modern flyer or menu card
  • + Vibrant use of shapes and accents to separate categories
  • + Excellent food photography detail within the grid
  • Text is largely gibberish and poorly rendered
  • Section headings like 'PIESMANETS' and '/MEANS' are incoherent
  • Layout feels a bit cluttered at the bottom

Wan 2.5 (Preview)

  • + Successfully followed all sector instructions including Appetizers, Pizza, and Mains
  • + Extremely clean and minimalist aesthetic with excellent use of negative space
  • + Text is much more legible and logically placed next to corresponding items
  • The grid is slightly uneven across the different rows
  • Several menu items use the exact same image assets
  • Minor spelling errors like 'Menue'

Verdict: Wan 2.5 (Preview) produced a superior menu by adhering more closely to the requested sections and maintaining a cleaner, more readable minimalist layout. Qwen Image 2512 had high-quality individual photos but failed significantly on text legibility and logical organization, resulting in a layout that feels more like a generic flyer template than a functional menu.

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.”

Qwen Image 2512
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2512

  • + Excellent photorealistic texture on the meat patty and vegetables
  • + Accurate rendering of the starburst price tag
  • + Very clear and legible typography with a realistic flame effect
  • The burger feels less 'exploded' as some layers are still resting flat on each other
  • The background is more static with less smoke and atmospheric depth

Wan 2.5 (Preview)

  • + Stronger sense of motion and 'exploded' dynamics with pieces flying outward
  • + Beautiful integrated typography with molten/dripping fire effects
  • + Better atmospheric smoke and lighting coherence
  • The meat patty texture is slightly less realistic compared to the other image
  • The price tag lacks the secondary 'glowing, fiery' text effect requested

Verdict: Both models followed the prompt exceptionally well. Qwen Image 2512 provides superior realism in the food textures and price starburst, while Wan 2.5 (Preview) offers a more creative and dynamic composition with better integration of text and atmospheric fire effects. Wan 2.5 (Preview) is the likely winner for better capturing the 'exploded' motion and the stylized 'MAGIC' aesthetic requested.

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.”

Qwen Image 2512
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2512

  • + Excellent adherence to the 'elegant cursive' requirement for the title.
  • + Highly legible text with very few spelling errors (only 'Risitto').
  • + Provides a realistic chalkboard aesthetic with natural dust and smudges.
  • Internal spelling error: 'Risitto' instead of 'Risotto'.
  • The prices look slightly more like digital overlays compared to the items.

Wan 2.5 (Preview)

  • + Great 'chalky' texture with smudges and varying stroke weight.
  • + Correct spelling of 'Risotto'.
  • + Dynamic layout with underlines that feels authentic to a café setting.
  • Missed the 'elegant cursive' requirement for the title, using a more blocky font.
  • Text repetition at the bottom ('$9' appears twice for the final item).
  • Awkward line break and spacing for the 'Grilled Octopus' entry.

Verdict: Qwen Image 2512 followed the stylistic instructions much better, specifically capturing the requested elegant cursive title. While it had a minor spelling error, Wan 2.5 (Preview) failed the cursive requirement and contained repetitive text ($9 twice) on the final menu item.

The Reversed Rodeo

Text-to-Image

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

Qwen Image 2512
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2512

  • + Excellent high-resolution skin and fabric textures.
  • + Very sharp focus on the subjects with realistic lighting on the space suit.
  • Completely failed the negative constraint/spatial instruction of 'horse on top'.
  • Anatomical issues with the horse's front legs being fused or poorly positioned.

Wan 2.5 (Preview)

  • + Dynamic composition with a sense of motion and cinematic lighting.
  • + Includes more 'surreal' elements like the galaxy and dust trails.
  • Failed the negative constraint/spatial instruction of 'horse on top'.
  • Distorted hoof anatomy and blurry foreground textures.

Verdict: Both Qwen Image 2512 and Wan 2.5 (Preview) failed the primary spatial challenge of placing the horse on top of the astronaut, both reverting to the standard trope of an astronaut riding a horse. Qwen Image 2512 is preferred for its superior technical clarity and realistic textures, whereas Wan 2.5 (Preview) suffers from more anatomical distortions and a softer focus.

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.”

Qwen Image 2512
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2512

  • + Natural composition with a focus on the characters through the front windshield
  • + Accurate depiction of a capybara's fur texture and calm expression
  • + Good adherence to the 'bored expression' prompt for the passenger
  • The capybara's hands look overly human/primate-like rather than rodent paws

Wan 2.5 (Preview)

  • + Excellent depiction of a rainy Times Square atmosphere with vibrant bokeh
  • + High level of detail on the capybara and the taxi's exterior/interior elements
  • + The capybara's paws look more anatomically appropriate for the species
  • The passenger's posture is slightly awkward with the seat positioning
  • Text on the taxi sign is nonsensical

Verdict: Both models followed the prompt exceptionally well, capturing the surreal humor of a capybara driving a taxi while a passenger remains indifferent. Wan 2.5 (Preview) is the winner due to its superior lighting, atmospheric rain details, and more realistic anatomical rendering of the capybara's paws compared to Qwen Image 2512.

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.”

Qwen Image 2512
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2512

  • + Includes all the requested text details accurately
  • + Effective cinematic lighting on the jack-o-lantern
  • + Strong gothic atmosphere with symmetrical composition
  • Spelling error in primary title ('Hallowern')
  • Background is slightly dark, making some thorns hard to distinguish

Wan 2.5 (Preview)

  • + Perfect text rendering for all requested copy
  • + High visual quality with detailed textures on the pumpkin and parchment
  • + Excellent adherence to the 'vintage parchment' and 'dark thorns' request
  • The parchment cutout shape is a bit abrupt against the dark background

Verdict: Wan 2.5 (Preview) is the winner because it followed the text instructions perfectly, including complex titles and the small scroll banner, without any spelling errors. While Qwen Image 2512 provides a moody and thematic layout, the misspelling of 'Halloween' in the main header makes it unusable for an invitation.

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.”

Qwen Image 2512
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2512

  • + Excellent typography with stylized outlines and drop shadows consistent with the cartoon theme.
  • + Highly detailed and appealing miniature scene with varied sushi types and garnishes.
  • + Stronger 3D diorama aesthetic with a clearly defined square base.
  • The composition is slightly off-center compared to the text.
  • Shadows on the base are a bit soft, losing some of the isometric crispness.

Wan 2.5 (Preview)

  • + Perfectly centered composition and clean, modern typography.
  • + Excellent soft-lighting and subsurface scattering effects on the sushi rice and fish.
  • + Ultra-clean visual quality with zero artifacts.
  • Much simpler scene with only one piece of nigiri compared to model A's variety.
  • The flag icon is slightly disconnected from the text block's visual style.

Verdict: Qwen Image 2512 offers a more complete miniature diorama with varied sushi types and better stylized typography that fits the 'cartoon' prompt perfectly. Wan 2.5 (Preview) produces a cleaner, more minimalist image with superior lighting, but feels less like a full 'scene'.

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.”

Qwen Image 2512
Wan 2.5 (Preview)
0% wins 0% ties 100% wins

AI Judge Analysis

Qwen Image 2512

  • + Excellent fur texture and realistic lighting integration
  • + Clearer, more naturalistic animal faces and eyes
  • + Higher consistency in anatomical structure
  • The animals are static and sitting rather than 'chasing and tumbling' as requested
  • The central puppy appears disproportionately large compared to the fox

Wan 2.5 (Preview)

  • + Successfully captures the dynamic 'chasing' and 'tumbling' action from the prompt
  • + Beautiful inclusion of dew sparkles and floating seeds/pollen
  • + Vibrant colors with a strong sense of movement
  • The fox kit has strange, glowing blue eyes that look unnatural
  • Resolution on the kitten's face is slightly blurry and lacks the fine detail seen in the other model
  • Anatomical issues where the dog's back leg/tail area is confusingly rendered

Verdict: Qwen Image 2512 produces a much more polished and photorealistic portrait with superior fur detail, but it misses the dynamic action requested in the prompt. Wan 2.5 (Preview) does a better job of capturing the joyful movement and 'tumbling' energy, though it suffers from lower clarity and some strange artifacts, particularly in the fox's eyes. Qwen is the preferred choice for visual quality, while Wan is better for thematic adherence.

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.”

Qwen Image 2512
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2512

  • + Excellent typography with correct accents and stylish calligraphy
  • + High-quality illustrative shading and texture
  • + Strong adherence to the 'vintage emblem' style
  • The steam is a bit heavy, leanly more towards 'illustration' than 'minimalist logo'

Wan 2.5 (Preview)

  • + Successfully captures a more minimalist, clean vector look
  • + Accurate text rendering and layout on banners
  • + Good use of positive and negative space in the cloche
  • The cloche looks more like a candle jar or bell jar than a food cloche
  • The background texture includes a distracting 'crumpled paper' effect that creates harsh lines

Verdict: Qwen Image 2512 provides a much more professional and aesthetically pleasing design that feels like a premium heritage brand. While Wan 2.5 (Preview) follows the 'minimalist' instruction better, its execution of the cloche is less recognizable, and the background artifacts detract from the overall quality. Qwen Image 2512 is the better choice for its artistic depth and superior typography.

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.”

Qwen Image 2512
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2512

  • + Successfully includes the requested Saturn V icon and Lunar Module surface landing icon.
  • + Good use of the specified NASA color palette.
  • + Includes names of the astronauts in a logical grouping.
  • The text contains numerous spelling errors and hallucinations like 'Translaurtoit' and 'Desceeint'.
  • The numbering of the steps is repetitive and non-sequential (two #2s, two #3s).
  • A prompt instruction 'Steps stop at landing' was mistakenly rendered as billboard text.

Wan 2.5 (Preview)

  • + Excellent text rendering with no spelling errors for the main steps.
  • + Superior layout that uses lines to show the logical flow of the mission trajectory.
  • + Strong adherence to the flat-vector infographic style with clean, minimalist iconography.
  • One of the icons shows a space shuttle-style craft instead of a lunar module for the descent phase.
  • The 'Armstrong' and 'Aldrin' portraits look extremely similar/cloned.
  • One moon icon is placed oddly within the trajectory line.

Verdict: Wan 2.5 (Preview) produces a far more professional infographic with readable, correctly spelled text and a logical layout that follows the mission's trajectory. While Qwen Image 2512 includes more detailed lunar module icons, its severe text hallucinations and broken numbering sequence make it unusable as an infographic.

Next steps

Explore each model