Head to head
Esc

Models · slot A

to navigate to pick

FLUX.2 [dev] Black Forest Labs Wan 2.7 Alibaba

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

FLUX.2 [dev]

24.5 arena score

#18 of 62 in Text-to-Image

Skill signature · Text-to-Image

Wan 2.7

20.5 arena score

#39 of 62 in Text-to-Image

Vote tally

Where the votes landed

FLUX.2 [dev]

0%

win rate

Ties

0%

Wan 2.7

0%

win rate

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent photographic lighting and depth of field.
  • + Perfect adherence to spatial instructions with the sphere inside the cube.
  • + High realism in textures, particularly the wood grain and glass reflections.
  • The sphere appears slightly glossier than expected, bordering on a marble texture.

Wan 2.7

  • + Realistic vintage look for the red book with visible binding details.
  • + Accomplishes all prompt requirements including lighting direction.
  • + Natural-looking houseplant with believable leaf structures.
  • Perspective and reflection errors; there is a phantom blue sphere reflection/artifact on the left side of the glass.
  • The glass cube has internal vertical seams that make it look more like a case than a solid object.

Verdict: Both models followed the prompt perfectly in terms of object placement and basic attributes. FLUX.2 [dev] is the winner because it handles the physics of light and reflections much more accurately, whereas Wan 2.7 contains a strange ghostly reflection of the sphere on the left panel that doesn't correspond to the sphere's actual position.

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Successfully preserved the man's specific clothing and hairstyle
  • + Maintained the interior details and leather of the car
  • + Excellent integration of the man into the driver's seat with realistic lighting
  • The exterior of the car is cropped out of the frame
  • The man's facial features changed slightly compared to the source

Wan 2.7

  • + Accurately captured the full vista of the California coastline
  • + Preserved the full exterior of the car correctly
  • + Good sense of motion on the road
  • Completely failed to use the specific man from the source image
  • The person in the car is a generic, low-detail figure
  • Lost the likeness and outfit requested in the image editing prompt

Verdict: FLUX.2 [dev] successfully performed the complex task of merging the two source images, placing the specific man in his specific clothes inside the car. While the car's exterior is partially cut off, the character preservation is excellent. Wan 2.7 failed to use the source man at all, instead generating a generic scene that ignores the identity of the person provided.

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent adherence to the motion blur requirement for passing cars.
  • + Very realistic skin texture and age-appropriate details on the hands and face.
  • + Superior lighting and reflections on the wet jacket and pavement.
  • The bicycle geometry is slightly confusing with excess frame bars and overlapping parts.
  • The 'imperfect framing' is very subtle, remaining fairly centered.

Wan 2.7

  • + Strong candid 'street photo' composition with a wide-angle feel.
  • + Captures the wet pavement reflections and rainy atmosphere very well.
  • + Great clothing texture showing dampness and wear.
  • Completely missed the 'motion blur from passing cars' instruction; the cars are crisp.
  • Hand-to-handlebar interaction is slightly muddy and less detailed than Image A.
  • The skin texture is a bit smoother and less 'natural' than Model A.

Verdict: FLUX.2 [dev] followed the complex prompt requirements much more closely, specifically capturing the motion blur of passing cars and providing a much more intimate, shallow depth of field. While Wan 2.7 has a nice street photography feel, it failed the specific technical instruction regarding motion blur and the skin textures are less detailed.

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent depiction of colorful beads in the braids
  • + Beautiful warm torchlight reflection and glowing bokeh sparks
  • + High quality engraving details on the chest plate
  • The scars look a bit like fresh red stage makeup rather than weathered battle scars
  • The torch in the corner is a bit blurry and takes up significant frame space

Wan 2.7

  • + Exceptional skin texture with realistic pores and believable battle-worn scars
  • + Highly detailed braided hair and leather-like texture on the straps
  • + Strong sense of 'battle-worn' through the dented and scratched armor
  • The beads are very uniform and simple compared to Model A
  • The torchlight is slightly less warm on the skin tones

Verdict: FLUX.2 [dev] excels in color and lighting, providing a very cinematic look with vibrant beads and warm reflections. However, Wan-2.1 produces a more convincing 'battle-worn' character with superior skin texture, realistic scabs/scarring, and weathered armor that matches the prompt's gritty requirements more effectively.

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent grid layout with clear horizontal and vertical alignment.
  • + Strong, bold typography used for headings that is easy to read.
  • + Accurate categorization of appetizers, pizza, and mains.
  • The text content within the menu items is mostly gibberish.
  • The food photos lack descriptive labels making it feel like a template shell.

Wan 2.7

  • + Features actual legible text and coherent food labels.
  • + High visual appeal with a photorealistic presentation including props around the menu.
  • + Consistent iconography and a clean, modern aesthetic with brand elements like a QR code.
  • The 'sides' and 'drinks' section photos include a burger and cake, which feels inconsistent with the header navigation.
  • Small text becomes slightly blurry upon close inspection.

Verdict: Wan 2.7 is the clear winner for its high level of detail, inclusion of legible English text, and realistic presentation. While FLUX.2 [dev] creates a strong structural layout that mimics a traditional menu fold, the content is nonsensical, whereas Wan 2.7 provides a complete, usable design concept with appropriate food pricing and branding.

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent photorealistic texture on the meat patty and bun
  • + Clean and readable text with a consistent fiery glow effect
  • + Perfect vertical alignment of the exploded burger components
  • The 'exploded' effect is quite subtle compared to model B
  • The sauce droplets look slightly artificial

Wan 2.7

  • + Very dynamic composition with a strong sense of motion and 'exploded' elements
  • + Detailed texture on the lettuce and grilled patty
  • + Creative fiery typography for the main title
  • Sesame seeds are floating individually in space rather than being on the bun
  • The 'starburst' for the price looks more like a sun or gear than a traditional graphic burst

Verdict: FLUX.2 [dev] produces a more professional and photorealistic advertisement with superior text integration and clean lighting. While Wan 2.7 has a more creative and 'exploded' composition, it suffers from odd artifacts like floating sesame seeds and less realistic material rendering.

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent chalk texture throughout the board.
  • + Strong adherence to the requested 'handwritten' style with visible smudges and imperfections.
  • + Successfully completed the truncated third item correctly based on context.
  • The spacing on the second item 'Grilled Octopus' is a bit cluttered.
  • The handwriting style is slightly messy, though fits the 'chalkboard' aesthetic.

Wan 2.7

  • + Very clean and legible text alignment.
  • + Good composition with decorative divider lines.
  • + Successfully completed the third item as 'Brown Butter Chocolate Chip Cookies'.
  • The text looks more like a digital font than actual hand-drawn chalk.
  • Fails the prompt's requirement for 'realistic chalk handwriting' and 'no digital fonts'.
  • The 'chalk' texture on the letters is too uniform and lacks natural variation.

Verdict: FLUX.2 [dev] significantly outperformed Wan 2.7 by following the specific stylistic constraints of the prompt. While Wan 2.7 produced legible text, it looked like a digital overlay, whereas FLUX.2 [dev] successfully captured the smudged, gritty, and variable texture of real chalk handwriting on a board.

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Successfully integrated character elements like the sunglasses, scarf, and black sweater.
  • + Captured the likeness of the man from Image 2.
  • Horrific anatomical failure with a second head and distorted limbs.
  • Failed to match the pose correctly, resulting in a surreal and incoherent composition.

Wan 2.7

  • + Maintained perfect anatomical coherence and image quality.
  • + Preserved the environment and lighting of the source pose image perfectly.
  • Completely failed the character swap instruction.
  • Made no visible changes to the person from Image 1.

Verdict: This comparison represents a complete failure in different directions. FLUX.2 [dev] attempted the edit but produced a disturbing, nightmarish image with multiple heads and broken anatomy, while Wan 2.7 failed to perform any edit at all, simply returning a cropped version of Image 1. FLUX.2 [dev] is technically the winner for attempting the prompt, though the result is unusable.

The Reversed Rodeo

Text-to-Image

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

FLUX.2 [dev]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent cinematic lighting and atmosphere
  • + Realistic textures on both the horse's coat and the spacesuit
  • + Dynamic and natural horse pose
  • Failed the specific spatial instruction for the horse to be on top of the astronaut

Wan 2.7

  • + Clear and sharp focus throughout the image
  • + Consistent anatomy for the horse and astronaut
  • + Well-defined background elements like galaxies and planets
  • Failed the specific spatial instruction for the horse to be on top of the astronaut
  • Composition feels slightly flat compared to Model A

Verdict: Both FLUX.2 [dev] and Wan 2.7 failed the subversion task of placing the 'horse on top' of the astronaut, instead defaulting to the common trope of an astronaut riding a horse. FLUX.2 [dev] is the superior image due to its cinematic lighting, richer colors, and more realistic textures, whereas Wan 2.7 has a more digital, illustrative feel.

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Successfully transferred the specific blue coat and plaid scarf from Image 2.
  • + Maintained the sand patches on the person's face.
  • + Good lighting integration on the fabric.
  • Failed to preserve the source person's face, blending it heavily with the face from Image 2.
  • Ignored the requirement to keep the background completely unchanged, adding a beach that wasn't there in the same way.

Wan 2.7

  • + Excellent preservation of the source person's identity, vitiligo patterns, and facial structure.
  • + Maintained the exact background and wood structure from Image 1.
  • + High-quality rendering of fabric and jewelry detail.
  • Failed to use the 'exact outfit' from Image 2, instead generating a generic 'elaborate' gold-trimmed coat and vest.
  • Distorted the hand details slightly.

Verdict: Both models failed the specific 'transfer' instruction in different ways: FLUX.2 [dev] successfully moved the clothing but failed to preserve the source person's identity, resulting in a face merge; Wan 2.7 perfectly preserved the source person and background but hallucinated a completely different elaborate outfit instead of using the one provided in Image 2. Wan 2.7 is the likely winner because it followed the 'source preservation' rules for the person and background much more effectively, even though it failed the specific fashion matching.

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent fur texture and lighting on the capybara.
  • + Accurately places the passenger in the back seat as requested.
  • + The bokeh in the background maintains a high degree of photorealism.

Wan 2.7

  • + Sharp details on the car exterior and taxi sign.
  • + Good depiction of the capybara's paws on the steering wheel.
  • The passenger is incorrectly sitting in the front passenger seat.
  • The capybara's fur has a slightly stylized, comb-like appearance rather than being photorealistic.
  • Anatomical issues with the capybara's paws looking more like human-monkey hybrids.

Verdict: FLUX.2 [dev] followed the spatial instructions perfectly, placing the businesswoman in the back seat and creating a highly convincing cinematic atmosphere. Wan 2.7 failed the prompt's layout request by putting the passenger in the front seat and produced a less realistic, more artificial-looking capybara.

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent typography with perfect spelling and alignment.
  • + High-contrast cinematic lighting that creates a moody, spooky atmosphere.
  • + Central Jack-o-lantern is well-integrated with the background elements.
  • The parchment effect is limited to the outer edges rather than the whole background.
  • The 'webs and thorns' border is a bit dark and blends into the background shadow.

Wan 2.7

  • + Successfully captures a 'vintage poster' aesthetic with detailed illustrations.
  • + Includes all requested text elements accurately along with creative additions.
  • + Complex and symmetrical composition with great use of the parchment texture.
  • The character of the central jack-o-lantern is slightly more cartoonish than 'cinematic'.
  • The image feels a bit crowded with many small secondary elements like skulls and books.

Verdict: FLUX.2 [dev] produces a more professional and atmospheric invitation with superior lighting and clean, elegant typography. While Wan 2.7 does a great job with the vintage parchment theme and includes many charming details, FLUX.2 [dev]'s focus on composition and cinematic quality makes it a more effective design for an actual event invitation.

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]
Before After
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent preservation of the original image's background, clothing, and facial features.
  • + Realistic hair texture and complex lighting integration.
  • The choice of an oversized afro style feels less 'natural' as an extension of the subject's existing beard and aesthetic.
  • The hairline integration at the forehead lacks some fine transition detail.

Wan 2.7

  • + Provides a very natural and believable hairstyle that fits the subject's age and existing beard perfectly.
  • + Flawless preservation of the original image's environment and lighting.
  • + The hairline is extremely realistic and well-integrated.
  • There are minor texture artifacts where the hair meets the top edge of the glasses frame.

Verdict: Both models perform exceptionally well at preserving the source image, keeping the background and clothing identical. Wan 2.7 is the winner because the chosen hairstyle is far more congruent with the subject's facial features and beard, whereas FLUX.2 [dev] opted for a disproportionately large afro that feels like a costume choice rather than a natural restoration.

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent adherence to the typographic layout requested.
  • + High-quality PBR material rendering on the sushi fish and rice.
  • + Cleaner, more minimalist diorama base that feels more premium.
  • The white plate is slightly off-center on the diorama base.
  • The text 'SUSHI' is relatively plain compared to the bold 'JAPAN'.

Wan 2.7

  • + Beautiful color palette and vibrant textures on the ingredients.
  • + Great variety of sushi items included in the miniature scene.
  • + The diorama base has a nice rounded architectural feel.
  • Failed to follow text placement instructions, putting text in a single block with the flag to the right.
  • Included many more elements (chopsticks, sauce, napkins) than the 'minimal' request.
  • The shrimp tail has a slight anatomical oddity with multiple segment points.

Verdict: FLUX.2 [dev] followed the layout instructions much more accurately, placing the text and flag exactly where requested while maintaining a clean, minimal aesthetic. Wan 2.7 produced a more visually 'busy' scene with more variety, but ignored the specific typographic hierarchy and text placement requested in the prompt.

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Expertly captures the caricature art style with exaggerated proportions.
  • + Integrated the TV anchor profession seamlessly into a hockey rink setting.
  • + Preserved the subject's likeness effectively within the stylized art form.
  • The text on the anchor desk is garbled and unreadable.
  • The inclusion of a second human caricature was not requested and feels a bit cluttered.

Wan 2.7

  • + Maintained the original 'selfie' composition from the source image.
  • + Humorous incorporation of hockey themes, such as the dog in the goalie mask and the tiny rink.
  • + Clear, readable text in the speech bubbles.
  • The facial caricature is a bit generic and loses some of the subject's specific likeness compared to the source.
  • The transition between the subject and the TV background is slightly messy.

Verdict: FLUX.2 [dev] produced a much more professional and traditional caricature that feels like a cohesive piece of editorial art, successfully blending all prompt elements into a single scene. While Wan 2.7 kept the selfie framing of the original image and added funny elements like the dog in a helmet, the artistic quality and likeness match in FLUX.2 [dev] are superior.

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent photorealism with soft, naturalistic lighting
  • + High detail in the fur texture and dew droplets
  • + Features all requested animals with clear, expressive eyes
  • Added an extra bunny that wasn't explicitly requested
  • The animals are sitting still rather than 'playfully chasing' or 'tumbling'

Wan 2.7

  • + Better captures the action of 'playfully chasing' and 'tumbling'
  • + Strong 'god rays' lighting effect as requested
  • + All four specific animals are present and uniquely identifiable
  • The fox looks slightly more like an adult than a kit
  • Anatomically awkward kitten paws and tail positioning
  • The lighting feels slightly more artificial and over-saturated than Image A

Verdict: Both models followed the prompt well, but FLUX.2 [dev] produced a much more photorealistic and aesthetically pleasing image with superior fur texture. While Wan 2.7 better captured the 'playful' movement requested in the prompt, it suffered from minor anatomical wonkiness and a less realistic finish compared to FLUX.2 [dev].

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent adherence to the 'Ghibli' style with clean linework and soft, flat coloring.
  • + Successfully transforms the background into a dreamy, meadow-like setting requested in the prompt.
  • + Perfectly captures the 'warm, nostalgic' lighting and pastel color palette.
  • Changes the background entirely, losing the original urban street context.

Wan 2.7

  • + Successfully preserves the original composition, including the background buildings and passersby.
  • + The watercolor texture aligns well with the 'hand-painted' requirement.
  • + Retains the recognizable facial features of the people in the original meme.
  • Art style feels more like a standard Western watercolor illustration than Studio Ghibli.
  • The lighting lacks the 'dreamy' quality requested in the prompt.

Verdict: FLUX.2 [dev] delivers a much more convincing Studio Ghibli aesthetic, perfectly capturing the distinctive eye shapes, soft coloring, and whimsical background elements typical of that style. While Wan 2.7 does a better job of preserving the source image's specific background details, its artistic style is less accurate to the specific 'Ghibli' request, appearing instead as a generic watercolor sketch.

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]
Before After
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent depiction of wind-blown hair with high energy
  • + Perfect preservation of the original person and dog's identity
  • + Adds a significant amount of leaves to enhance the motion effect
  • The hair takes on a slightly unnatural, fan-like spread
  • Some leaves overlap the subjects in a slightly cluttered way

Wan 2.7

  • + More natural and subtle hair movement that matches a real breeze
  • + Leaves have motion blur which enhances the 'dynamic' feel
  • + Maintains the lighting and colors of the source image perfectly
  • Fewer leaves added compared to Model A, feeling slightly less 'energetic'
  • The motion effect on the hair is less dramatic than requested

Verdict: Both models succeeded at adding motion while preserving the source image perfectly. FLUX.2 [dev] provides a more literal and dramatic interpretation of 'energetic' motion with a large volume of hair and leaves, while Wan 2.7 offers a more realistic, photographically accurate representation including motion blur on the flying debris.

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent typography with perfect spelling of 'Caffè Florian'.
  • + Beautifully minimalist vector style that adheres perfectly to the 'emblem' request.
  • + Superior handling of the banner and subtle paper texture.
  • The steam is very simple, though this fits the minimalist style.

Wan 2.7

  • + Ornate and attractive circular frame design.
  • + Clean, readable text for the 'Est. 1720' portion.
  • + Good interpretation of the 'warm brown' color palette.
  • Failed spelling of the primary subject, rendering the name as 'Florion' instead of 'Florian'.
  • Lack of 'banner' for the establishment date as requested; it uses a badge instead.
  • Less 'minimalist' than requested, bordering on cluttered.

Verdict: FLUX.2 [dev] followed every instruction perfectly, including the specific spelling of the restaurant and the minimalist vector aesthetic. Wan 2.7 failed the spelling of 'Florian' and incorporated too many extra elements (stars, leaves) that contradicted the 'minimalist' requirement.

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]
Wan 2.7

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent vector art style with high-quality gradients and shading
  • + Accurate depiction of the Lunar Module and Saturn V components
  • + High clarity in individual icons
  • Poor layout logic with random grid placement
  • Garbled text in several labels
  • Failed to follow the numbered step sequence requested

Wan 2.7

  • + Perfect logical layout following the requested step-by-step sequence
  • + Excellent textual accuracy and supporting data integration
  • + Strong adherence to the 'infographic' format with a clear flow
  • Icon design is more simplistic compared to Model A
  • One minor typo ('Descript' instead of 'Descent')
  • Visual elements are slightly more cramped

Verdict: While FLUX.2 [dev] produces much higher quality individual vector illustrations, it fails completely as an infographic, providing a jumbled grid with illegible text. Wan 2.7 successfully interprets the prompt's intent for a structured, educational poster, following the mission steps in order and including impressively accurate supporting data and readable text.

Next steps

Explore each model