Head to head
Esc

Models · slot A

to navigate to pick

HiDream I1 Full HiDream AI Wan 2.5 (Preview) Alibaba

Settled by community votes across 16 shared challenges, with an AI judge weighing in on each.

HiDream I1 Full

15.2 arena score

#60 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

HiDream I1 Full

0%

win rate

Ties

0%

Wan 2.5 (Preview)

0%

win rate

Shared challenges 16

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.”

HiDream I1 Full
Wan 2.5 (Preview)

AI Judge Analysis

HiDream I1 Full

  • + Excellent depiction of the plant visible through the glass refraction
  • + Higher realism in the glass texture and reflections
  • + Balanced and cinematic lighting composition
  • The blue sphere is levitating unrealistically in the center of the cube
  • The table edge is cropped closely

Wan 2.5 (Preview)

  • + Natural placement of the blue sphere resting on the bottom surface
  • + Detailed texture on the vintage red book cover
  • + Stronger sense of depth with the wider table view
  • The blue sphere has an odd reflection or phantom object at the top of the cube
  • Noticeable dust/particle noise artifacts in the air

Verdict: Both models followed the complex spatial instructions perfectly. HiDream I1 Full is the winner due to its superior rendering of glass refraction and reflections, creating a more believable photographic scene, whereas Wan 2.5 (Preview) has distracting noise artifacts and a strange visual glitch appearing between the book and the sphere.

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.”

HiDream I1 Full
Wan 2.5 (Preview)

AI Judge Analysis

HiDream I1 Full

  • + Excellent shallow depth of field effect.
  • + High-quality reflections on the wet pavement.
  • + Atmospheric lighting that feels cinematic.
  • The man appears to be sitting through the bicycle frame rather than repairing it.
  • Misses the 'motion blur from passing cars' request as cars look relatively sharp.
  • Anatomy issues where the man's leg and the bike frame merge.

Wan 2.5 (Preview)

  • + Stronger narrative with the man actually using tools to repair the bike.
  • + More realistic skin textures and fine details on the hands.
  • + Better adherence to the 'light rain' prompt with visible raindrops hitting surfaces.
  • The bicycle kickstand and pedals are geometrically confusing and physically impossible.
  • Lack of requested 'motion blur' from background cars.
  • Composition feels a bit tight, missing the 'imperfect framing' candid feel.

Verdict: Both models failed to include the requested motion blur for passing cars, but Wan 2.5 (Preview) produced a far more logical and detailed scene where the subject is actively engaged in repairing the bike. HiDream I1 Full has a beautiful cinematic atmosphere and reflections, but the subject's interaction with the bicycle is physically broken, with limbs and frame pieces merging into an incoherent mess.

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.”

HiDream I1 Full
Wan 2.5 (Preview)

AI Judge Analysis

HiDream I1 Full

  • + Excellent depiction of ornate, engraved plate armor textures.
  • + Strong adherence to the 'beads in hair' request with complex braids.
  • + Intense, lifelike eye contact and sharp facial detail.
  • The scars appear somewhat artificial, like paint or surface marks rather than deep skin tissue.
  • The lighting on the face feels a bit flat compared to the dramatic background.

Wan 2.5 (Preview)

  • + Outstanding lifelike skin texture with realistic dirt and varied scarring.
  • + Exceptional detail on the worn leather straps and frayed cloth underlayer.
  • + Natural and moody lighting with warm torchlight reflections.
  • The character looks quite young for a 'battle-worn' paladin.
  • The armor engravings are slightly less sharp than Model A.

Verdict: While HiDream I1 Full produces a very heroic and sharp image with impressive armor, Wan 2.5 (Preview) better captures the specific atmospheric details of the prompt. Wan 2.5's rendering of leather, frayed cloth, and realistic skin texture creates a more believable and 'lifelike' scene 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.”

HiDream I1 Full
Wan 2.5 (Preview)

AI Judge Analysis

HiDream I1 Full

  • + Strong, legible bold sans-serif text
  • + High-quality, realistic food photography
  • + Clean minimalist layout with ample white space
  • Repetitive food imagery (all pizza for different sections)
  • Incorrect header grouping (Appetizers/Pizza combined)
  • Garbled placeholder text for descriptions

Wan 2.5 (Preview)

  • + Excellent adherence to the grid layout request
  • + Diverse and colorful food photos corresponding to different sections
  • + Clever use of vibrant color-coded accents for the sections
  • Text is largely illegible and contains typos like 'Restormalit Menue'
  • Some food items are repetitive or logically misplaced (pizza in mains)

Verdict: Wan 2.5 (Preview) produced a much more comprehensive and visually interesting layout that followed all prompt requirements, including the grid structure and vibrant accents. HiDream I1 Full had superior text rendering and photo realism but failed to provide variety in the food images, using four nearly identical pizzas for every section.

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.”

HiDream I1 Full
Wan 2.5 (Preview)

AI Judge Analysis

HiDream I1 Full

  • + Features a clean and aesthetically pleasing wood frame
  • + Captures an 'ordered' menu vibe with columns
  • Fails the prompt instructions for specific menu items entirely
  • Text is mostly nonsensical gibberish
  • The 'chalk' texture looks more like a digital font than actual handwriting

Wan 2.5 (Preview)

  • + Excellent prompt adherence, correctly rendering the specific dishes and prices requested
  • + Highly realistic chalk texture with dust smudges and varying stroke thickness
  • + Captures a convincing handwriting style as requested
  • The perspective makes some words near the bottom slightly harder to read
  • Minor repetitiveness in the price for the third item

Verdict: Wan 2.5 (Preview) significantly outperforms HiDream I1 Full by actually following the text instructions and menu items provided in the prompt. While HiDream I1 Full generates a visually clean image, its text is almost entirely illegible gibberish, whereas Wan 2.5 (Preview) renders complex phrases and realistic chalk effects with high accuracy.

The Reversed Rodeo

Text-to-Image

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

HiDream I1 Full
Wan 2.5 (Preview)

AI Judge Analysis

HiDream I1 Full

  • + Strong cinematic lighting with high contrast
  • + Good sense of motion in the horse's tail and mane
  • + Accurately depicts the astronaut riding on top of the horse as requested
  • The horse's hind legs are anatomically confused and merge into the dust
  • The composition feels a bit cramped at the edges

Wan 2.5 (Preview)

  • + Excellent anatomical detail on the horse and spacesuit
  • + Vibrant, high-resolution textures and cosmic background
  • + Dynamic composition with a sense of scale provided by the nebula and Earth
  • The astronaut's legs and the saddle integration look a bit awkward
  • The 'surreal' aspect is less pronounced compared to the more grounded cinematic look

Verdict: Both models successfully followed the specific instruction to put the astronaut on top of the horse, avoiding a common AI pitfall. Wan 2.5 (Preview) produced a significantly higher quality image with better skin/fur textures and a more expansive, detailed background. HiDream I1 Full captured a nice cinematic mood, but struggled with the anatomy of the horse's legs.

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.”

HiDream I1 Full
Wan 2.5 (Preview)

AI Judge Analysis

HiDream I1 Full

  • + Excellent texture on the capybara's fur and clothing.
  • + The capybara's hands/paws are anthropomorphized convincingly on the wheel.
  • + Accurate text rendering on the hat.
  • The passenger's scale seems slightly off compared to the driver.
  • The perspective from the window makes the car feel a bit cramped.

Wan 2.5 (Preview)

  • + Strong composition showing both the interior and exterior of the taxi with rain details.
  • + Accurately captures the 'bored' expression of the businesswoman.
  • + Great atmospheric lighting consistent with Manhattan at night.
  • The paws on the steering wheel look like wrinkled human fingers or claws rather than capybara paws.
  • The taxi sign on top of the car has nonsensical text/symbols.

Verdict: HiDream I1 Full produces a much higher quality capybara with realistic fur texture and a more professional-looking outfit, though the interior feels less like a real New York taxi. Wan 2.5 (Preview) captures the requested atmosphere and character expressions better, including the bored businesswoman and rainy city lights, but fails significantly on the rendering of the paws and the text on the taxi signage. HiDream I1 Full is the winner for its superior technical execution of the main subject.

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.”

HiDream I1 Full
Wan 2.5 (Preview)

AI Judge Analysis

HiDream I1 Full

  • + Strong composition with a clear central focal point
  • + Effective vintage graphic design style
  • Failed significantly on text rendering, producing gibberish for the date and location
  • Missing the requested 'thorns' in the border element

Wan 2.5 (Preview)

  • + Excellent text rendering, accurately including all specific event details and the scroll banner
  • + Rich, cinematic lighting and detailed textures on the pumpkin and background
  • + Incorporated both webs and thorns into the border as requested
  • The 'Halloween' text is slightly warped along a curve
  • The scroll banner is small relative to the main heading

Verdict: Wan 2.5 (Preview) followed the prompt instructions much more accurately, particularly regarding the specific text details like the date, time, and location which HiDream I1 Full failed to render correctly. Wan 2.5 also provided a much more polished and cinematic visual experience with the requested thorns and lighting effects.

Bald man challenge

Image Editing
Edit instruction

“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”

Before After
HiDream I1 Full
Before After
Wan 2.5 (Preview)

AI Judge Analysis

HiDream I1 Full

  • + Attempts to add hair length in the back.
  • Completely changes the background, facial features, and clothing.
  • Adds bizarre, nonsensical artifacts like the object protruding from the beard.
  • Fails to provide a 'full thick head of hair' as the top remains bald.

Wan 2.5 (Preview)

  • + Perfectly preserves all original image elements including face, lighting, and background.
  • + Provides a very realistic, thick head of hair that matches the beard texture.
  • + Seamlessly integrates the new hair with the existing facial structure and glasses.

Verdict: Wan 2.5 (Preview) executed the edit perfectly, providing a natural and thick head of hair while keeping every other detail of the source image identical. HiDream I1 Full failed the task entirely, generating a new image with different features, a different background, and nonsensical visual artifacts while leaving the top of the head bald.

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.”

HiDream I1 Full
Wan 2.5 (Preview)

AI Judge Analysis

HiDream I1 Full

  • + Excellent adherence to the isometric miniature 3D cartoon style
  • + Superior variety and detail in the sushi models on a realistic diorama plate
  • + Clean text rendering for both words
  • Failed to include the requested small flag icon
  • The 'JAPAN' text is slightly cut off at the top corners

Wan 2.5 (Preview)

  • + Successfully included all prompt elements including the small flag icon
  • + High visual clarity with soft, professional lighting and textures
  • + Perfectly centered composition
  • The scene is very minimalist with only one piece of sushi, while the prompt suggested a miniature 3D cartoon scene
  • Angle is more eye-level 3D rather than the requested 45° top-down isometric view

Verdict: HiDream I1 Full delivers a much more interesting and detailed isometric diorama that matches the requested 45-degree angle perfectly, though it misses the flag icon. Wan 2.5 (Preview) follows the literal text instructions better by including the flag, but the scene feels sparse and the perspective is not truly isometric as requested.

Over-the-top cartoon caricature

Editing
Edit instruction

“Create a caricature of me and my job. Make it exaggerated and humorous, incorporating my profession as a tv show anchor and my love for dogs and hockey.”

Source
HiDream I1 Full
Wan 2.5 (Preview)

AI Judge Analysis

HiDream I1 Full

  • + Maintains a stronger facial resemblance to the source image
  • + High-quality digital painting style with good shading
  • + Cleverly includes hockey players on the TV screen in the background
  • The 'hockey' element is very subtle and nearly missed
  • The dog's placement on the shoulder is a bit awkward and lacks clear context
  • Fails to clearly depict the 'tv show anchor' profession beyond a generic headset

Wan 2.5 (Preview)

  • + Excellent adherence to all prompt elements: anchor desk, microphone, hockey stick, and multiple dogs
  • + True caricature style with classic proportions (large head, small body)
  • + Very creative interpretation, including a dog wearing a hockey jersey
  • The facial resemblance back to the source image is weaker than Model A
  • The hand holding the microphone has anatomical issues (six fingers/distorted grip)

Verdict: Model B (Wan 2.5 (Preview)) is the clear winner for prompt adherence, as it successfully integrated the TV anchor desk, a hockey stick, a playback screen showing hockey, and multiple dogs into a cohesive caricature composition. While Model A (HiDream I1 Full) produced a more polished digital painting that looks more like the subject, it was too subtle with the requested career and hobby elements.

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.”

HiDream I1 Full
Wan 2.5 (Preview)

AI Judge Analysis

HiDream I1 Full

  • + Strong composition with a clear focal point and warm, aesthetic lighting.
  • + High level of fur detail on the puppy and fox.
  • Failed to include the baby bunny, instead providing two kittens.
  • The image has a noticeably AI-generated, 'illustrated' feel rather than hyper-photorealism.
  • The poses are static rather than 'tumbling' or 'chasing' as requested.

Wan 2.5 (Preview)

  • + Included all four requested animals: puppy, kitten, bunny, and fox.
  • + Captures the action of the prompt much better with jumping and running poses.
  • + Excellent rendering of lighting effects, including god rays and sparkling dew drops.
  • The fox kit has unnatural, overly saturated blue/glowing eyes.
  • Some anatomy issues, specifically the kitten's front paw and the puppy's floating mid-air position.

Verdict: Wan 2.5 (Preview) is the clear winner as it successfully included all four requested animals and captured the dynamic, playful movement described in the prompt. While HiDream I1 Full produced a very cute image, it missed the bunny entirely and provided a static portrait rather than a scene of animals tumbling and chasing butterflies.

Studio Ghibli Anime Style

Editing
Edit instruction

“Transform this photo into a Studio Ghibli–inspired illustration. Use soft pastel colors, hand-painted textures, gentle lighting, dreamy backgrounds, and a warm, nostalgic mood”

Source
HiDream I1 Full
Wan 2.5 (Preview)

AI Judge Analysis

HiDream I1 Full

  • + Successfully captures a whimsical, vibrant anime aesthetic.
  • + Effective use of hand-painted textures and warm lighting.
  • + Creative interpretation of the characters in a new setting.
  • Fails to preserve the core composition and identities from the source image.
  • Crucial narrative element—the 'distracted boyfriend' dynamic—is lost as characters look in different directions.

Wan 2.5 (Preview)

  • + Excellent source preservation, maintaining the exact poses and layout of the original meme.
  • + Accurately applies a Studio Ghibli-inspired art style while keeping the scene recognizable.
  • + Faithfully replicates clothing details like the specific plaid pattern on the shirt.
  • The facial expressions are slightly softened compared to the intensity of the original photo.

Verdict: Wan 2.5 (Preview) is the clear winner for its superior ability to perform an image-to-image transformation while preserving the source's composition and narrative. While HiDream I1 Full creates a pleasant illustration, it ignores the structural layout of the source image, whereas Wan 2.5 perfectly translates the iconic meme into the requested Ghibli style.

Golden Hour Stroll

Image Editing
Edit instruction

“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”

Before After
HiDream I1 Full
Before After
Wan 2.5 (Preview)

AI Judge Analysis

HiDream I1 Full

  • + Excellent sense of motion and energy
  • + Effective hair-blow effect
  • Failed to preserve the source image subjects
  • Removed the dog entirely from the scene
  • Changed the identity of the person

Wan 2.5 (Preview)

  • + Almost perfect source preservation of the person and dog
  • + Applied all requested edits while keeping the original composition
  • + Natural-looking wind effect on hair
  • The flying leaves look slightly flat and pasted on compared to the background

Verdict: Wan 2.5 (Preview) successfully performed the edit while maintaining the integrity of the original photo, keeping the woman's face and the dog consistent. HiDream I1 Full completely ignored the source preservation aspect, regenerating a new person and removing the dog, essentially failing as an edit tool despite the good aesthetic quality of the new 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.”

HiDream I1 Full
Wan 2.5 (Preview)

AI Judge Analysis

HiDream I1 Full

  • + Clean vector aesthetic
  • + Accurate rendering of the banner text 'Est. 1720'
  • + Elegant inclusion of a coffee cup icon
  • Completely missed the primary logo text 'Caffè Florian'
  • The cloche is opaque rather than the traditional transparent look
  • Banner shape is slightly stiff and disconnected

Wan 2.5 (Preview)

  • + Successfully included all requested text including 'Caffè Florian'
  • + Attractive paper-like background texture
  • + Better interpretation of a cloche dome with transparency and depth
  • Slightly more complex than a 'minimalist' vector might be
  • Text alignment on the bottom banner is slightly off-center

Verdict: Wan 2.5 (Preview) followed the prompt instructions much more closely by including both required text elements, whereas HiDream I1 Full completely omitted the restaurant name. Wan 2.5 (Preview) also delivered a more sophisticated illustration style that better captures the 'vintage' and 'cloche dome' themes.

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.”

HiDream I1 Full
Wan 2.5 (Preview)

AI Judge Analysis

HiDream I1 Full

  • + Strong artistic vector style with clean lines.
  • + Captures the NASA-inspired color palette perfectly.
  • + Large, readable main title text.
  • Fails to include more than 4 steps of the mission.
  • Includes nonsensical text like 'UNG' and 'UNLY'.
  • Iconography is somewhat generic and doesn't follow a logical mission flow.

Wan 2.5 (Preview)

  • + Successfully follows the logical flow of the 6 requested steps.
  • + Renders accurate and recognizable Apollo 11 mission icons, including the Lunar Module.
  • + Includes creative extra details like the astronaut portraits and 'Tranquility' base location.
  • Text layout for 'Descent' and 'Landing' is floating and disconnected from the icons.
  • Misses the primary main title at the top of the poster.

Verdict: Wan 2.5 (Preview) is the clear winner as it provides an actual infographic that follows all 6 steps of the requested sequence, whereas HiDream I1 Full only provides a loosely related collage of images with garbled text. Although HiDream I1 Full has a slightly cleaner vector aesthetic, Wan 2.5 (Preview) demonstrates much higher prompt adherence by including the specific icons and names requested for the moon landing sequence.

Next steps

Explore each model