Head to head
Esc

Models · slot A

to navigate to pick

FLUX.2 [dev] Flash fal Wan 2.7 Pro Alibaba

Settled by community votes across 20 shared challenges, with an AI judge weighing in on each.

FLUX.2 [dev] Flash

27.1 arena score

#5 of 62 in Text-to-Image

Skill signature · Text-to-Image

Wan 2.7 Pro

20.7 arena score

#38 of 62 in Text-to-Image

Vote tally

Where the votes landed

FLUX.2 [dev] Flash

100.0%

win rate

Ties

0.0%

Wan 2.7 Pro

0.0%

win rate

100.0% 0.0% ties 0.0%
Shared challenges 20

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.”

FLUX.2 [dev] Flash
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent photographic realism and lighting consistency.
  • + Matches every detail of the prompt accurately.
  • + Highly coherent glass physics and reflections.
  • The plant is slightly less detailed in the background compared to image B.

Wan 2.7 Pro

  • + Beautifully detailed fern in the background.
  • + Good use of natural light and texture on the wooden table.
  • The glass cube has internal vertical glass dividers that were not requested.
  • The reflection of the blue sphere on the bottom of the cube appears more like a mirror than simple glass.

Verdict: FLUX.2 [dev] Flash perfectly adheres to the prompt while maintaining a extremely high level of realism, especially in how the glass cube is constructed. Wan 2.7 Pro produces a lovely image but introduces unnecessary structural elements inside the glass cube that distract from the main subject.

Man and Car in California

Editing
Edit instruction

“Make a photo of the man driving the car down the California coastline”

Source
FLUX.2 [dev] Flash
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Successfully merges the man and car from source images with high fidelity.
  • + Retains the man's specific hairstyle, plaid coat, and black scarf perfectly.
  • + Excellent interior detail preservation for the car's cream seats and dashboard.
  • The perspective of the road behind the car is slightly inconsistent with the car's angle.
  • Changes the car's interior dashboard to wood grain which wasn't clearly visible in the source.

Wan 2.7 Pro

  • + Stays very faithful to the original car's exterior design and lighting.
  • + Captures the iconic California coastline aesthetic with high visual quality.
  • + Retains the exact wheel design and front grille detail from the source.
  • Fails to include the specific man from the source image, replacing him with a generic figure.
  • The person in the car is much smaller and less detailed than the man in the source image.

Verdict: FLUX.2 [dev] Flash is the clear winner because it successfully followed the complex instruction to merge two specific subjects: the man from the second image and the car from the first. Wan 2.7 Pro preserved the car well but completely ignored the identity of the person provided in the source image, substituting a generic man instead.

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.”

FLUX.2 [dev] Flash
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent adherence to motion blur requests on passing cars
  • + Realistic skin texture and age details on the man's hands and face
  • + Superior handling of reflections and wet pavement aesthetics
  • The structural anatomy of the bicycle is slightly muddled where the man is working
  • The background cars look a bit dated for a modern street photo

Wan 2.7 Pro

  • + Realistic bicycle geometry and clear reflection in the puddle
  • + Accurate shallow depth of field
  • Failed to include motion blur on the passing cars as requested
  • The lighting on the man feels slightly flat compared to the background
  • The rain effect is very subtle, almost invisible compared to Model A

Verdict: FLUX.2 [dev] Flash followed the prompt much more closely, successfully incorporating the requested motion blur and cinematic atmosphere. While Wan 2.7 Pro produced a very clean image with a realistic bicycle, it failed to execute the specific camera effects like motion blur and had a much less convincing 'light rain' atmosphere.

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.”

FLUX.2 [dev] Flash
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent adherence to the 'beads in braids' prompt with vibrant, colorful details.
  • + Highly detailed etched armor texture and leather strap grain.
  • + Effective cinematic symmetry and realistic facial injuries.
  • The facial scars appear more like fresh bloody cuts rather than 'faint scars'.
  • The bead placement on the forehead hair feels slightly unnatural.

Wan 2.7 Pro

  • + Subtle, realistic scarring and skin texture that perfectly matches 'faint scars'.
  • + Warm torchlight reflections are very well integrated into the armor highlights.
  • + Strong depth-of-field effect with high-quality bokeh.
  • Armor engraving is slightly softer and less defined than Model A.
  • The beads are clustered at the top of the head rather than integrated throughout the braids.

Verdict: FLUX.2 [dev] Flash provides a more striking, detailed image with impressive textures on the armor and accessories, though it makes the character look more actively wounded than 'battle-worn'. Wan 2.7 Pro excels at the subtle facial details and lighting requested in the prompt, offering a more realistic interpretation of 'faint scars'. Ultimately, FLUX.2 [dev] Flash is the winner due to its superior sharpness and detail in the armor, leather, and braided hair.

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.”

FLUX.2 [dev] Flash
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent vibrant color accents that match the prompt.
  • + High-quality, appetizing pizza photography within the grid.
  • + Good adherence to the bold sans-serif font request for headers.
  • Nonsensical text and spelling errors in the headers like 'ASSLAL5' and 'RIZLA'.
  • Visual confusion between sections as all food images are pizza regardless of the header name.

Wan 2.7 Pro

  • + Outstanding text legibility and logical menu item descriptions.
  • + More realistic professional layout including a QR code, social media icons, and a footer.
  • + Excellent variety in food photos including appetizers, mains, and drinks.
  • Color accents are limited to orange and brown rather than the 'vibrant accents' requested.
  • Some minor spelling errors in small body text like 'Tirusu' or 'Riautto'.

Verdict: Wan 2.7 Pro is the clear winner because it functions as a realistic menu design with legible text, diverse food imagery, and a coherent logical structure. While FLUX.2 [dev] Flash followed the 'vibrant' color prompt more closely, its content was repetitive (all pizzas) and the text was largely gibberish.

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.”

FLUX.2 [dev] Flash
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent typography with a realistic fiery glowing effect
  • + Highly photorealistic textures on the bun and meat patty
  • + Dynamic composition with realistic sauce droplets and embers
  • The layout is a bit bottom-heavy with the large starburst overlapping the background

Wan 2.7 Pro

  • + Very symmetrical and clean layout for a social media ad
  • + Good variety of ingredients including pickles and onions not explicitly named in the prompt
  • + Clean and readable text labels
  • The 'fiery' effect on the text and starburst looks like a flat graphic mask rather than realistic light
  • The lighting on the ingredients is a bit inconsistent and looks more like a digital composite than a single photograph

Verdict: FLUX.2 [dev] Flash produces a much more professional and photorealistic result, particularly with the integrated fiery text effects which truly look like they are glowing. While Wan 2.7 Pro offers a very clean ad-style layout, its graphics and lighting feel more like clip-art compared to the cinematic quality of FLUX.2 [dev] Flash.

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.”

FLUX.2 [dev] Flash
Wan 2.7 Pro
100% wins 0% ties 0% wins

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Natural chalk texture with realistic dusty residue
  • + Excellent variety in letter sizing and hand-drawn qualities
  • + Highly accurate text rendering relative to the prompt
  • The 'Brown Butter' line repeats the price twice ($9 and $9) inappropriately
  • The background café elements are slightly more blurred than Model B

Wan 2.7 Pro

  • + Crisp and legible handwriting style
  • + Clean composition with nice wood-paneled background
  • + Followed the text content perfectly without repetition
  • Text looks more like a digital font than natural chalk handwriting
  • The 'elegant cursive' request for the title was not fully met, as it is mostly print-style capitals
  • Lacks the fine chalk dust and pressure variations present in Model A

Verdict: FLUX.2 [dev] Flash captures the requested chalk texture and authentic handwriting style much more effectively, although it hallucinates a redundant price on the final line. Wan 2.7 Pro produces much cleaner, perfect text, but it lacks the natural variations and realistic 'chalk on board' feel specified in the prompt, resulting in a look that feels more like a digital font.

Pose & Character Mashup

Editing
Edit instruction

“Use Image 1 as the exact pose reference and Image 2 as the character reference. Recreate the person/character from Image 2 in the exact dynamic pose and body position from Image 1. Keep the exact face, hair, clothing style/details, and expression from Image 2. Match the lighting and environment of Image 1. The final image must show the character from Image 2 performing the precise action/pose from Image 1 with perfect anatomy and natural integration.”

Source
FLUX.2 [dev] Flash
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Successfully integrated clothing, accessories, and facial features from the character reference.
  • + Correctly matched the yellow background and red ottoman from the pose reference.
  • + Maintained the lighting consistency of the scene.
  • Failed to replicate the dynamic, contorted pose from the source image accurately.
  • Significant anatomical artifacts, including a phantom head/hair appearing behind the main character.
  • The pose is static and crouching rather than the specific lean-and-reach of the reference.

Wan 2.7 Pro

  • + Perfectly preserved the pose and environment from Image 1.
  • Completely failed the character transfer instruction by leaving the original woman in the image.
  • Ignored all details from the character reference (Image 2) including gender, clothing, and accessories.

Verdict: This was a complex image-to-image task that both models struggled with. FLUX.2 [dev] Flash attempted the character and clothing transfer with high fidelity to Image 2, but it failed to correctly map those attributes onto the specific pose of Image 1, resulting in poor anatomy. Wan 2.7 Pro completely failed the instruction to change the character, simply returning the source pose image. FLUX.2 [dev] Flash is preferred as it actually attempted the edit and successfully transferred the character's identity and attire.

The Reversed Rodeo

Text-to-Image

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

FLUX.2 [dev] Flash
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + High cinematic quality with realistic atmospheric lighting.
  • + Excellent texture details on the horse's fur and the astronaut's suit.
  • Failed the specific spatial instruction (astronaut is riding the horse).

Wan 2.7 Pro

  • + Clarity in composition with a wide variety of celestial bodies.
  • + Clean rendering of the astronaut and horse anatomy.
  • Failed the specific spatial instruction (astronaut is riding the horse).
  • The visual style is less cinematic and more like a collage compared to the prompt requirements.

Verdict: Both FLUX.2 [dev] Flash and Wan 2.7 Pro failed to follow the specific 'horse on top' instruction, which is a known difficulty for current text-to-image models. FLUX.2 [dev] Flash is preferred because it produced a much more cinematic and cohesive image with superior lighting and atmospheric detail, whereas Wan 2.7 Pro feels somewhat flatter and less integrated.

Outfit Transfer Challenge

Editing
Edit instruction

“Use Image 1 as the base person. Dress them in the exact elaborate outfit from Image 2 (including all layers, accessories, jewelry, and shoes). Carefully adapt the clothing to the body shape and pose in Image 1 while maintaining realistic fabric behavior, correct proportions, and perfect lighting/shadow matching. Keep the person’s exact face, hair, and background completely unchanged.”

Source
FLUX.2 [dev] Flash
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Successfully replicates the blue coat and plaid scarf from Image 2.
  • + Captures various accessories like necklaces and rings.
  • Catastrophic failure in facial preservation, resulting in a distorted, multi-faced monster effect.
  • The person's background is modified to include a strange secondary head/hairline behind the main subject.

Wan 2.7 Pro

  • + Excellent preservation of the original person's face, hair, and vitiligo patterns.
  • + Fits the clothing naturally to the subject's pose and full body.
  • Fails to use the specific outfit from Image 2, substituting it with a different ornate blazer.
  • Removes the shirt entirely, which was not requested.

Verdict: This is a comparison of two failures in instruction following. FLUX.2 [dev] Flash captures the correct clothing from Image 2 but completely destroys the base person's face, making the image unusable. Wan 2.7 Pro preserves the person and composition perfectly but ignores the specific outfit provided in the reference image, opting for a generic gold-patterned jacket instead. Wan 2.7 Pro is the winner simply for maintaining the structural integrity of the human subject, despite the inaccuracy in fashion selection.

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.”

FLUX.2 [dev] Flash
Wan 2.7 Pro
100% wins 0% ties 0% wins

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent photorealistic texture on the capybara fur and clothing.
  • + Correctly places the businesswoman in the back seat as requested.
  • + The capybara's paws on the steering wheel look natural and well-integrated.
  • The passenger is technically in the passenger seat next to the driver instead of the 'back seat'.
  • The text on the taxi sign 'TAX' is slightly garbled.

Wan 2.7 Pro

  • + Very high detail in the capybara's face and paws.
  • + Dynamic perspective from the side of the vehicle.
  • The businesswoman is sitting in the front passenger seat, not the back seat.
  • The paws on the steering wheel look more like primate hands than capybara paws.
  • The human's hands holding the phone have anatomical irregularities.

Verdict: Both models failed to place the businesswoman in the back seat, placing her in the front passenger seat instead. FLUX.2 [dev] Flash is the winner because it provides a more realistic rendering of capybara paws and a more convincing 'bored' expression on the passenger, whereas Wan 2.7 Pro generated hands that look more like a monkey's and had artifacts on the human's hands.

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.”

FLUX.2 [dev] Flash
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent typography with a cohesive gothic font choice
  • + Cinematic lighting that creates a deep, spooky atmosphere
  • + Accurate rendering of the requested text including dates and locations
  • Includes a small line of garbled 'extra' text (Burk: 03696) above the time
  • The parchment effect is less 'vintage poster' and more like a dark illustration

Wan 2.7 Pro

  • + Strong composition with a clear 'vintage parchment' border design
  • + Extremely clean and accurate text rendering for all requested fields
  • + Creative additions like the cauldron and crow that enhance the Halloween theme
  • The pumpkin's internal lighting (a candle on a stick inside) looks slightly physically incoherent
  • The 'glow' is less atmospheric compared to the other model

Verdict: Both models performed exceptionally well on typography and prompt adherence. FLUX.2 [dev] Flash has superior cinematic lighting and atmosphere, but Wan 2.7 Pro produces a much clearer 'invitation' layout with better border details and perfect text without any hallucinations.

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
FLUX.2 [dev] Flash
Before After
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Successfully adds a very dense and full head of hair.
  • + Preserves the original facial features and environment accurately.
  • The hair texture and style (afro-textured) do not match the subject's existing beard and eyebrow texture, appearing unrealistic.
  • The hairline transition is a bit harsh.

Wan 2.7 Pro

  • + Excellent texture matching with the existing beard and eyebrow hair.
  • + Very realistic and natural hairline integration.
  • + Flawless preservation of the source image's lighting and facial structure.
  • None notable.

Verdict: Wan 2.7 Pro provided a much more realistic result by matching the hair texture to the subject's existing beard, whereas FLUX.2 [dev] Flash generated a texture that felt disconnected from the subject's features. Wan 2.7 Pro also achieved a more natural-looking hairline and better overall integration.

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.”

FLUX.2 [dev] Flash
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent adherence to the '45° top-down isometric' and 'raised diorama base' instructions.
  • + Very clean, bold typography and a perfectly rendered flag icon.
  • + High-quality PBR textures, particularly on the wooden base and fish.
  • The sushi pieces are slightly merged or awkwardly stacked compared to traditional plating.
  • Shadows on the base are a bit soft, losing some of the 'ultra-clean' isometric edge.

Wan 2.7 Pro

  • + Beautifully varied sushi types with vibrant colors and excellent material definition.
  • + Crisp typography that includes an extra tagline with clear legibility.
  • + Highly polished 3D cartoon aesthetic with great attention to detail in the garnish.
  • Ignored the request for a solid light blue background by adding floating artifacts and extra props like chopsticks.
  • The camera angle is a standard perspective rather than the requested 45-degree isometric top-down view.

Verdict: FLUX.2 [dev] Flash followed the technical layout instructions much more closely, delivering the specific isometric 45-degree angle and diorama base requested. While Wan 2.7 Pro produced more visually appealing and diverse sushi assets, it failed the composition constraints by adding unrequested elements and using a standard perspective instead of isometric.

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
FLUX.2 [dev] Flash
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent preservation of the subject's facial features in a caricature style.
  • + Cleverly integrates hockey and dogs by making the dogs hockey players.
  • + High-quality, polished illustration style with consistent lighting.

Wan 2.7 Pro

  • + Successfully captures the requested themes with clear hobby-related icons.
  • + The caricature is very expressive and humorous.
  • The facial resemblance to the source image is significantly lower than Model A.
  • Includes bizarre elements like a microphone attached to a golf club/hockey stick hybrid.
  • The speech bubbles are a bit generic and the background is less detailed.

Verdict: FLUX.2 [dev] Flash is the clear winner as it maintains a much stronger facial resemblance to the source woman while masterfully blending the themes of hockey, dogs, and news anchoring into a cohesive rink-side studio. Wan 2.7 Pro creates a generic character with strange props like a microphone-stick hybrid and has lower overall artistic polish.

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.”

FLUX.2 [dev] Flash
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent depiction of dew sparkles and golden sunrise rays
  • + Clean, high-quality rendering of fur and facial expressions
  • + Stronger adherence to the '8K masterpiece' aesthetic with sharp focus
  • Added extra animals (two foxes, two bunnies) not specified in the count
  • The composition feels slightly more posed than a natural action scene

Wan 2.7 Pro

  • + Captures the 'playfully chasing' and 'tumbling' movement very effectively
  • + Correct quantity of each animal type mentioned in the prompt
  • + Beautifully rendered golden hour lighting and god rays
  • The fox's face appears slightly warped or less 'baby-like' compared to the others
  • The kitten's pose and paws are a bit awkward and anatomically weak

Verdict: While FLUX.2 [dev] Flash produces a cleaner and more visually polished image with better dew details, it fails on basic count instructions by doubling up on several animals. Wan 2.7 Pro adheres strictly to the list of subjects and better captures the requested 'playful chasing' motion, despite slightly lower anatomical precision in the fox and kitten.

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
FLUX.2 [dev] Flash
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Stronger adherence to the Studio Ghibli art style with characteristic facial features and eye shapes.
  • + Successfully incorporates a 'dreamy' background with soft sunsets and floral elements as requested.
  • + Excellent application of soft pastel colors and gentle lighting.
  • Replaced the urban background with a meadow, which significantly changes the original setting.
  • The addition of a small rabbit creature in the background might be unnecessary for some users.

Wan 2.7 Pro

  • + Preserves the original source image's urban background layout much more accurately.
  • + High-quality watercolor textures that feel like hand-painted concept art.
  • + Maintains the distinct facial expressions of the three people perfectly.
  • The art style leans more toward generic watercolor illustration rather than a distinct 'Ghibli' look.
  • The lighting is less 'warm and nostalgic' compared to the other model.

Verdict: FLUX.2 [dev] Flash captures the Ghibli aesthetic much more effectively through its stylized character design and dreamy lighting, though it sacrifices the original urban background to do so. Wan 2.7 Pro preserves the original composition and environment significantly better, but its 'Ghibli' interpretation is less pronounced, looking more like a standard watercolor sketch. FLUX.2 is the winner for better following the complex stylistic requirements of the prompt.

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
FLUX.2 [dev] Flash
Before After
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Successfully adds a large number of falling leaves to create a wind-blown atmosphere.
  • + Effectively edits the hair to spread outward symmetrically.
  • + Preserves the details of the woman, dog, and background with high fidelity.
  • The hair flow looks slightly stiff and symmetrical, appearing a bit unnatural.
  • Some leaves appear very small and sharp, lacking motion blur to sell the 'dynamic' feel.

Wan 2.7 Pro

  • + Excellent hair modification that looks more natural and randomized than Model A.
  • + Leaves are varied in size and placement, contributing to a more organic feel.
  • + The lighting and overall energetic vibe feel more cohesive.
  • Slight loss of facial detail compared to the original source image.
  • Fewer leaves overall compared to Model A, though they are better integrated.

Verdict: Both models performed the edit well, but Wan 2.1 Pro is the winner because the hair movement feels much more realistic and the overall composition feels more 'energetic' as requested. While FLUX.1 [dev] Flash added more leaves, it resulted in a more static, symmetrical look for the hair that didn't quite capture the motion 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.”

FLUX.2 [dev] Flash
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent typography with correct spelling and accent mark.
  • + Clean emblem composition that fits the 'minimalist' request well.
  • + Accurate rendering of a cloche dome with simple steam lines.
  • The 'Est. 1720' text is slightly off-center under the banner.
  • The texture is very subtle, perhaps a bit too clean for a 'vintage' look.

Wan 2.7 Pro

  • + Beautifully detailed engraving-style illustration of the cloche.
  • + Rich vintage aesthetic with a cohesive circular emblem design.
  • + Includes nice extra details like the 'Venezia Italia' subtitle.
  • Spelling error in the main brand name ('Florion' instead of 'Florian').
  • The composition includes cluttered elements in the corners that detract from the logo focus.

Verdict: FLUX.2 [dev] Flash followed the instructions much more accurately, specifically providing the correct spelling for 'Caffè Florian' and maintaining a minimalist aesthetic. While Wan 2.7 Pro produced a very attractive vintage illustration, the spelling error and extra background elements made it less successful as a functional logo design.

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.”

FLUX.2 [dev] Flash
Wan 2.7 Pro

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Includes detailed illustrations of the Saturn V and Lunar Module.
  • + Good text legibility for the main headings.
  • + High-quality rendering of Earth and Moon textures.
  • Poor layout organization that doesn't follow a logical timeline.
  • Text garbling on certain labels like 'LAUNCH' and 'MOON'.
  • Mismatched icons (e.g., using a Space Shuttle-style rocket for Saturn V).

Wan 2.7 Pro

  • + Excellent adherence to the 'flat-vector' style and infographic layout.
  • + Logically organized vertical timeline from launch to landing.
  • + Accurate NASA-inspired color palette and clean iconography.
  • Small font size for the supporting technical details is difficult to read.
  • Minor spelling error in 'DESCRIPT' (meant to be Descent/Descent Stage).
  • The astronaut icons are very generic compared to the rest of the illustration.

Verdict: Wan 2.7 Pro followed the requested infographic structure much better, creating a logical vertical flow that tells the story of the mission. FLUX.2 [dev] Flash produced more detailed individual illustrations, but failed as an infographic due to a chaotic layout and several text errors.

Next steps

Explore each model