Head to head
Esc

Models · slot A

to navigate to pick

Qwen Image 2.0 Alibaba Wan 2.5 (Preview) Alibaba

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

Qwen Image 2.0

21.7 arena score

#34 of 62 in Text-to-Image

Skill signature · Text-to-Image

Wan 2.5 (Preview)

23.4 arena score

#26 of 62 in Text-to-Image

Vote tally

Where the votes landed

Qwen Image 2.0

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.”

Qwen Image 2.0
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2.0

  • + Excellent depiction of the glass cube with realistic internal reflections.
  • + High photo-realism regarding textures of the red book and wooden table.
  • + Accurately represents all requested spatial relationships.
  • The blue sphere appears to be floating without a physical support inside the cube, which feels slightly unnatural.
  • The glass cube contains some confusing internal planes that don't perfectly align with a simple 6-sided cube.

Wan 2.5 (Preview)

  • + Beautiful lighting with strong directional shadows and 'dust' particles for atmosphere.
  • + Good depth of field, creating a professional photography aesthetic.
  • + The blue sphere has a nice matte finish that contrasts well with the glass.
  • The red book is clipping awkwardly through the top pane of the glass cube.
  • The blue sphere is physically merged with the bottom pane of glass rather than resting on it or floating inside.

Verdict: Qwen Image 2.0 provides a more logically coherent scene with very sharp, realistic textures, although the floating sphere is a minor physical curiosity. Wan 2.5 (Preview) has a more artistic and moody lighting setup, but it fails on structural logic, specifically where the book and sphere intersect/clip with the glass cube surfaces.

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 2.0
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2.0

  • + Excellent depiction of skin texture and age spots
  • + More effective shallow depth of field and background bokeh
  • + Realistic 'imperfect framing' that matches a candid street photo
  • The bike chain and pedal assembly have some structural clipping issues
  • Motion blur of the passing car is quite subtle

Wan 2.5 (Preview)

  • + Stronger visual atmosphere with visible raindrops and reflections
  • + Very high resolution and clean composition
  • + Includes mechanical tools as a logical detail
  • The bike stands on a nonsensical, floating kickstand support
  • Skin texture looks slightly too smooth or 'AI-rendered' compared to Model A
  • The framing is too centered and clean, missing the 'imperfect' request

Verdict: Qwen Image 2.0 captures the requested 'candid street photo' aesthetic much more effectively with its tighter, slightly off-kilter framing and highly realistic skin textures. While Wan 2.5 (Preview) creates a beautiful, moody atmosphere with better rain effects, it feels more like a staged digital artwork and suffers from significant anatomical errors in the bicycle's kickstand and support 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.”

Qwen Image 2.0
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2.0

  • + Excellent realism in the skin texture, wrinkles, and battle-worn features.
  • + Very ornate and detailed engraving on the plate armor with logical reflections.
  • + Detailed rendering of the braided hair and colored beads.
  • The glowing orange eyes look somewhat supernatural/artificial rather than 'lifelike'.
  • The bokeh sparks in the foreground are a bit large and distracting from the face.

Wan 2.5 (Preview)

  • + Highly lifelike and expressive eyes with subtle reflections.
  • + Outstanding texture on the leather straps, buckles, and frayed cloth underlayer.
  • + Beautiful lighting from the visible torch, creating a natural warm-to-cool gradient on the face.
  • The character appears very young, which slightly clashes with the 'battle-worn' description compared to Model A.
  • The beads in the hair look more like metallic studs than traditional beads.

Verdict: Qwen Image 2.0 captures the 'battle-worn' aspect with more maturity and grit, featuring incredibly detailed armor and facial weathering. However, Wan 2.5 (Preview) produces a more balanced and aesthetically pleasing composition with superior lighting, more realistic eyes, and better execution of the specified textile and leather textures. While Qwen feels more like an epic fantasy still, Wan 2.5 feels more like a high-end cinematic photograph.

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 2.0
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2.0

  • + Excellent high-resolution food photography that looks appetizing and professional.
  • + Clean grid layout with large, clear section headers.
  • + Consistently bold and legible pricing typography.
  • The text descriptions under the food are heavily mangled and illegible.
  • Visual interpretation of the grid does not perfectly align with the category headers above it.

Wan 2.5 (Preview)

  • + Includes comprehensive menu elements like descriptions, headers, and organized color-coded sections.
  • + Good use of vibrant accents with red and green lines to separate categories.
  • + Better overall structural composition as a full-page menu design.
  • The food photos are repetitive and look quite similar across different categories.
  • Text rendering is poor with many typos and nonsense words.
  • Less realistic food photography compared to its competitor.

Verdict: Qwen Image 2.0 produces significantly better food photography with a more modern, high-end feel, though it struggles with the text descriptions. Wan 2.5 (Preview) does a better job of actually designing a multi-section menu with structural accents, but the repetitive nature of its food images makes it less effective for a professional presentation. Qwen Image 2.0 is the preferred choice for its superior visual quality and cleaner minimalism.

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 2.0
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2.0

  • + Excellent photorealistic texture on the patty and bun
  • + Highly effective fiery text effect that looks like actual flames
  • + Great use of steam and embers to add atmosphere
  • The burger is not really 'exploded' as most components are still stacked
  • The 'LIMITED TIME ONLY' text is small and slightly blurry

Wan 2.5 (Preview)

  • + Perfect adherence to the 'exploded' request with components widely suspended
  • + Text rendering is clean and follows the requested hierarchy well
  • + Very dynamic composition with splashing sauce and flying vegetables
  • The cheese has a slightly plastic, less photorealistic appearance
  • The 'MAGIC BURGER' text looks more like liquid/syrup than fire

Verdict: Wan 2.5 (Preview) followed the structural prompt much better, creating a truly 'exploded' and dynamic layout that feels like a professional ad. While Qwen Image 2.0 has superior photorealistic textures and more impressive flame effects on the text, it failed to separate the burger components in mid-air as 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 2.0
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2.0

  • + Excellent text rendering with no spelling errors.
  • + The handwriting looks authentic with realistic chalk smudges and texture.
  • + Perfect adherence to all prompt details including specific prices and dates.

Wan 2.5 (Preview)

  • + Strong visual composition with a wooden frame.
  • + High contrast text that is very easy to read.
  • + Accurate adherence to the requested menu items and date.
  • The text style looks more like a digital brush font than natural handwriting.
  • Minor formatting issues like the price appearing twice for the cookies ($9).
  • The 'elegant cursive' requirement for the title was not fully met, favoring a bold print style instead.

Verdict: Qwen Image 2.0 followed the prompt more effectively by delivering a truly handwritten chalk aesthetic with natural variations and realistic smudging. While Wan 2.5 (Preview) produced a clean and readable board, its text appears more like a digital font and contained a redundant price entry on the final item.

The Reversed Rodeo

Text-to-Image

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

Qwen Image 2.0
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2.0

  • + Excellent adherence to the 'surreal' instruction with scale-like patterns on the horse's skin.
  • + Clean rendering of the astronaut suit and horse features.
  • + Includes floating water droplets which enhance the zero-gravity cinematic feel.
  • The transition between the horse's back and the astronaut's legs is slightly clipped.
  • The star field is a bit repetitive in pattern.

Wan 2.5 (Preview)

  • + Dynamic composition with a beautiful nebula in the background.
  • + High quality textures on the horse's coat and mane.
  • + Realistic lighting and shadows on the astronaut and horse.
  • Failed the specific negative constraint 'horse on top, not vice versa' by placing the astronaut on top of the horse.
  • The dust/debris at the bottom implies a ground surface which slightly diminishes the 'in space' surrealism requested.

Verdict: Both models failed to follow the specific 'horse on top, not vice versa' logic flip, both producing the traditional astronaut-on-horse image. Qwen Image 2.0 is the winner as it captured the 'surreal' requirement better through the horse's unusual textures and the zero-gravity water effects, while Wan 2.5 (Preview) produced a more standard, ground-based looking trot albeit with a space background.

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 2.0
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2.0

  • + Excellent fur texture rendering
  • + Dynamic and realistic side-angle composition
  • + Captures the bored expression of the passenger perfectly
  • The passenger appears to be in the front seat or mid-cabin due to the car's perspective

Wan 2.5 (Preview)

  • + Clearer distinction between the front driver seat and back passenger seat
  • + Includes nice environment details like rain on the windshield and a taximeter
  • + Front-facing symmetrical composition is very clean
  • The capybara's 'paws' look more like human fingers or monkey hands
  • The capybara's expression is slightly more startled than calm/professional

Verdict: Both models followed the prompt well, but Qwen Image 2.0 captures a more authentic 'photorealistic' feel with superior lighting and texture on the capybara. While Wan 2.5 (Preview) handles the spatial arrangement of a taxi better, the anatomical rendering of the capybara's paws is unsettling compared to the more natural appearance in Qwen Image 2.0.

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 2.0
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2.0

  • + Excellent typography style that perfectly fits the gothic theme
  • + Flawless text rendering for all requested details
  • + Cohesive vintage parchment aesthetic throughout the entire frame

Wan 2.5 (Preview)

  • + Highly detailed, 3D-quality jack-o-lantern with fiery internal lighting
  • + Vibrant colors and high visual fidelity
  • + Complex layered border design with realistic thorns
  • The font choice for the main title is slightly generic and has inconsistent arched alignment
  • The transition between the central blue circular scene and the parchment background is a bit sharp

Verdict: Qwen Image 2.0 is the clear winner for this specific prompt because it better understands the 'vintage gothic poster' aesthetic, delivering perfectly legible and stylistically appropriate typography that feels integrated into the image. While Wan 2.5 (Preview) has impressive rendering quality and more dynamic lighting on the pumpkin, its text rendering and font choices feel less refined for a professional-looking 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 2.0
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2.0

  • + Excellent text legibility and accuracy
  • + High realism in textures, particularly the fish and wooden board
  • + Accurate representation of multiple types of sushi
  • Does not follow the 3D cartoon scene aesthetic, looking more like a photo
  • Text placement feels overlaid rather than integrated into a 3D environment
  • The wood grain is slightly warped on the right edge of the board

Wan 2.5 (Preview)

  • + Perfectly captures the 3D cartoon/stylized aesthetic requested
  • + High-quality PBR-style lighting and soft textures
  • + Excellent composition with a truly isometric feel and integrated text
  • Slightly lower resolution on the text edges
  • The Japanese flag icon is small compared to the text labels

Verdict: Wan 2.5 (Preview) better understood the stylistic requirements of the prompt, delivering a cohesive 3D cartoon miniature with appropriate PBR materials. Qwen Image 2.0 produced a much more realistic, photographic image which, while high quality, failed to capture the requested 'cartoon scene' and 'miniature' stylized aesthetic.

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 2.0
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2.0

  • + Excellent depiction of physical interaction and 'tumbling' described in the prompt
  • + Superior lighting and god ray integration that feels atmospheric
  • + Realistic proportions and textures for all four animals
  • The fox kit is a bit obscured at the bottom of the frame

Wan 2.5 (Preview)

  • + Very dynamic composition with all animals 'chasing' and moving toward the viewer
  • + Clearly includes all four animals with distinct features
  • + Vibrant colors and high clarity in the foreground flowers
  • The fox eyes look unnaturally blue and glassy
  • Floating water droplets look like digital overlays rather than natural dew
  • The kitten is missing a front paw/limb in its running pose

Verdict: Qwen Image 2.0 succeeds better as a 'hyper-photorealistic' masterpiece by creating a believable, heartwarming scene where the animals actually interact and play. Wan 2.5 (Preview) has a more energetic composition, but suffers from anatomical errors (the kitten's missing leg) and less realistic eye details for the fox.

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 2.0
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2.0

  • + Excellent typography rendering with the correct accent on 'Caffè'
  • + Strong vector illustration style with clean gradients
  • + Effective use of warm brown and cream tones
  • The placement of the main text inside the cloche is slightly unconventional for a logo
  • The 'steam' looks more like a flame icon

Wan 2.5 (Preview)

  • + Classic banner layout is more traditional for a vintage logo
  • + Good representation of a glass cloche with visible internal steam
  • + Sophisticated use of subtle paper texture in the background
  • The accent mark on 'Caffè' is missing or extremely faint
  • Composition feels a bit small relative to the total frame size

Verdict: Both models followed the prompt well, but Qwen Image 2.0 stands out for its superior typography and bold, clean vector execution. While Wan 2.5 (Preview) captured a more traditional logo layout and the specific 'steam' detail better, the crispness and text accuracy of Qwen Image 2.0 make it the more professional-looking graphic.

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 2.0
Wan 2.5 (Preview)

AI Judge Analysis

Qwen Image 2.0

  • + Strong vertical flow that logically guides the eye from launch to landing.
  • + Accurate typography with only one minor spelling error ('Translunjar').
  • + Clean, minimalist flat-vector style that perfectly matches the requested aesthetic.
  • The 'Descent' icon is a bit small and less detailed compared to the others.
  • Contains a typo in the word 'Translunjar'.

Wan 2.5 (Preview)

  • + Excellent iconography for the Saturn V and Lunar Module.
  • + Includes creative portait icons for the crew.
  • + High visual impact with a more complex arrangement of pathing lines.
  • Text layout is messy with 'TRANSLUNAR' placed away from its label group.
  • Incorrectly shows a Space Shuttle-style orbiter on top of the landing module path.
  • The step sequence is less intuitive and non-linear compared to Model A.

Verdict: Qwen Image 2.0 followed the structural requirements of the prompt more effectively, creating a clear and logical infographic despite a small typo. Wan 2.5 (Preview) produced higher-quality individual icons, particularly the crew and spacecraft, but failed on infographic coherence by including an unrelated shuttle-like craft and having a disjointed layout.

Next steps

Explore each model