Head to head
Esc

Models · slot A

to navigate to pick

FLUX.2 [dev] Black Forest Labs Wan 2.5 (Preview) Alibaba

Settled by community votes across 19 shared challenges, with an AI judge weighing in on each.

FLUX.2 [dev]

24.3 arena score

#19 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

FLUX.2 [dev]

0.0%

win rate

Ties

0.0%

Wan 2.5 (Preview)

100.0%

win rate

0.0% 0.0% ties 100.0%
Shared challenges 19

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Perfect adherence to all spatial instructions and objects.
  • + Highly realistic glass reflections and refractions of the background plant.
  • + Consistent and logical wood grain and lighting effects.
  • The plant in the background is slightly less sharp than the foreground elements.

Wan 2.5 (Preview)

  • + Excellent texture work on the red book cover and the plant leaves.
  • + Dynamic lighting with visible dust motes adds to the atmosphere.
  • The glass cube lacks a back panel or side faces, making it appear more like a glass frame or open structure.
  • The physics of the blue sphere's reflection on the bottom surface is inconsistent with its round shape.

Verdict: FLUX.2 [dev] followed the prompt more accurately by rendering a complete, enclosed glass cube with proper refraction of the plant behind it. While Wan 2.5 (Preview) produced beautiful textures and lighting, it failed to render the glass cube as a cohesive 3D object, resulting in a confusing internal structure.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent preservation of the man's specific hairstyle and clothing from the source image.
  • + Accurate reproduction of the car's interior details including the dashboard and white leather seats.
  • + High realism with cinematic lighting and a clear California coastline background.
  • The car's door and window line appear slightly altered or cut off compared to the original model.

Wan 2.5 (Preview)

  • + Successfully places the original car and a representation of the man into a wide-angle action shot.
  • + Great motion blur on the road and wheels conveying speed.
  • + Maintains the exterior profile of the car very well.
  • The scale of the man inside the car is significantly reduced, losing his identifiable facial features.
  • The man's clothing is not clearly preserved compared to the source image.

Verdict: FLUX.2 [dev] is the clear winner as it perfectly preserves the character's unique identity, clothing, and the car's interior details while placing them in the requested setting. Wan 2.5 (Preview) creates an appealing wide shot, but it fails to maintain the detail and likeness of the man from the source image.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent execution of the motion blur request for passing cars.
  • + Highly realistic skin texture and natural clothing details like raindrops on the jacket.
  • + Perfect adherence to the 'imperfect framing' and 'candid' street photography style.
  • The structure of the bicycle handlebars and brake lines is slightly Tangled/incoherent upon close inspection.

Wan 2.5 (Preview)

  • + Strong composition with clear reflections on the wet pavement.
  • + Good anatomical accuracy for the subject's posture and task.
  • + Beautiful shallow depth of field effect.
  • Failed to include the requested motion blur of passing cars.
  • The rain effect looks like static vertical lines rather than realistic droplets hitting surfaces.
  • Includes a strange artifacts where a kickstand and loose parts on the ground don't connect logically.

Verdict: FLUX.2 [dev] followed the technical requirements of the prompt much better, specifically capturing the motion blur of passing cars and the 'imperfect' candid framing that gives it a realistic street-photo feel. Wan 2.5 (Preview) produced a more 'staged' looking image that missed the motion blur requirement and had less realistic rain effects.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent depiction of ornate engraving on the plate armor
  • + Subtle but detailed hair braids with authentic-looking beads
  • + Naturalistic lighting from the torch source reflecting on the metal
  • The scars look a bit like surface-level scratches rather than healed or raw battle wounds
  • The bokeh sparks are a bit uniform in shape

Wan 2.5 (Preview)

  • + Strong texture on the frayed cloth underlayer and leather straps
  • + Compelling facial expression with tired, life-like eyes
  • + Excellent shallow depth of field with large, soft bokeh circles
  • The 'beads' in the hair look more like metal rivets or jewelry than traditional beads
  • The torch in the background is slightly incoherent with its holder

Verdict: Both models performed exceptionally well on this complex prompt. FLUX.2 [dev] stands out for its superior armor engraving and more natural hair/bead integration, while Wan 2.5 (Preview) excels in capturing the gritty texture of the cloth and a more emotive, battle-worn facial expression. FLUX.2 is the winner for its better adherence to the 'ornate engraved' and 'braided with beads' descriptions.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent grid layout with varied crops for food photography.
  • + Features legible prices and bold sans-serif headers.
  • + Clean, professional white background with a modern dual-column aesthetic.
  • Text consists of nonsensical characters and gibberish strings.
  • Repeating the 'Pizza' header on the second page is redundant.

Wan 2.5 (Preview)

  • + Beautiful, high-quality food photography with consistent lighting.
  • + Strict adherence to the 'Appetizers/Pizza/Mains' sections with a logical grid.
  • + Layout is more spacious and provides a better minimalist feel.
  • Text is highly garbled with many misspellings ('Restormalit Menue').
  • Top left text includes the meta-prompt description ('Modern minimalist') instead of a brand name.

Verdict: FLUX.2 [dev] creates a more realistic document structure that resembles a usable menu, though its text strings are entirely gibberish. Wan 2.5 (Preview) produces significantly better food imagery and a cleaner minimalist aesthetic, though it struggles with basic English spelling more than its competitor. Wan 2.5 is preferred for its superior artistic composition and adherence to the category requirements.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent text legibility and clean font choice
  • + Very photorealistic rendering of the meat patty and bun textures
  • + Consistent use of a glowing neon effect across all text elements
  • The 'exploded' effect is very static and vertically stacked compared to the prompt's request for motion
  • The cheese slice looks slightly artificial and flat

Wan 2.5 (Preview)

  • + Dynamic composition with a strong sense of motion and 'exploded' elements
  • + Highly creative typography with a dripping 'sauce' effect that fits the theme
  • + Rich atmosphere with smoke and floating sauce droplets enhancing the action
  • The 'LIMITED TIME ONLY' text is slightly warped
  • The background is less 'fiery' and more focused on smoke than Image A

Verdict: Both models followed the prompt exceptionally well, but Wan 2.5 (Preview) captured the 'dynamic' and 'exploded' aspect much more effectively with its angled composition and scattered ingredients. FLUX.2 [dev] produced a cleaner, more professional-looking advertisement with superior text rendering and photorealism, though it felt more like a static stack than a move-filled explosion.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent chalk texture with realistic dusty residue and streaks.
  • + Perfect adherence to the full menu text including the cut-off prompt portion.
  • + Very realistic handwriting style that looks authentically human.
  • The layout for the second item is slightly disjointed with the price floating.

Wan 2.5 (Preview)

  • + Clean, aesthetically pleasing composition with a wooden frame.
  • + High legibility and consistent cursive-adjacent style.
  • + Creative addition of underlines for emphasis.
  • Failed to include the full text for the second item, omitting 'Herbs'.
  • The text looks more like a digital font overlay than physical chalk on a board.
  • Repetitive price entry ($9) on the third item due to line breaks.

Verdict: FLUX.2 [dev] followed the text prompt much more accurately, correctly rendering 'Lemon & Herbs' and completing the final item name from the truncated prompt. FLUX.2 also achieved a superior 'chalk' texture that feels gritty and realistic, whereas Wan 2.5 (Preview) produced text that looks like a clean digital brush, missing the authenticity requested.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent anatomical realism in the horse's musculature and movement.
  • + High cinematic quality with realistic atmospheric lighting and subtle shadows.
  • + Logical composition with the horse galloping over the Earth's atmosphere.
  • The astronaut's hand position on the reins is slightly less defined.

Wan 2.5 (Preview)

  • + Vibrant, saturated colors in the nebula and space background.
  • + Clearer visor detail and distinct astronaut equipment highlights.
  • + Dynamic sense of motion with the dust and particles flying beneath the horse's hooves.
  • The horse's front left leg has a slight anatomical distortion where it connects to the shoulder.
  • The composition feels slightly more cluttered with high-contrast elements competing for attention.

Verdict: Both models successfully adhered to the complex prompt of an astronaut riding a horse in space. FLUX.2 [dev] is the winner due to its superior anatomical consistency and a more cohesive cinematic atmosphere, whereas Wan 2.5 (Preview) has a slight anatomical error in the horse's front leg despite its impressive vibrant colors.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Successfully transferred complex outfit elements including the scarf, sunglasses, and layered jewelry.
  • + Attempted to retain the unique vitiligo patterns on the forehead and neck areas.
  • + Maintained the beach background and wooden post accurately.
  • Failed to preserve the base person's face, instead blending it with the face of the man from Image 2.
  • Changed the person's hair texture and style significantly from the source image.

Wan 2.5 (Preview)

  • + High visual quality with clean integration of the coat and scarf.
  • + Preserved the background and beach setting perfectly.
  • Completely replaced the base person's face and hair with the man from Image 2, ignoring the instruction to keep the person unchanged.
  • Removed the unique vitiligo skin patterns which were a defining feature of the base person.
  • Omitted the jewelry layers requested in the clothing transfer.

Verdict: Both models failed the primary constraint of preserving the base person's face and hair, instead opting to face-swap or blend in the man from Image 2. However, FLUX.2 [dev] performed better as it attempted to retain the vitiligo skin patterns and successfully transferred the elaborate jewelry and accessories, whereas Wan 2.5 (Preview) produced a generic face swap that ignored the specific identity of the subject in Image 1.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent texture on the capybara's fur and the jacket fabric.
  • + Cinematic lighting that accurately reflects a city at night.
  • + Stronger adherence to the requested 'calm, professional expression' for the capybara.
  • The passenger's hands and phone interaction are slightly blurry.
  • The paws on the steering wheel look slightly anthropomorphized/distorted.

Wan 2.5 (Preview)

  • + Excellent composition with a wider view of the Times Square background.
  • + High level of detail on the car's exterior rain droplets and taxi sign.
  • + Great 'bored' expression on the businesswoman in the back seat.
  • The capybara's paws look more like human fingers with claws than actual capybara feet.
  • The taxi sign text 'C1B' is slightly nonsensical.

Verdict: FLUX.2 [dev] produces a more intimate, photorealistic shot with superior texture work on the subjects, creating a more convincing 'professional' look for the capybara. Wan 2.5 (Preview) offers a more vibrant composition with better environmental storytelling through the rain and Times Square lights, but falls slightly short on subject realism compared to FLUX.2.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent typography with perfect rendering of all requested text
  • + Highly cohesive gothic atmosphere with a dark, moody color palette
  • + Very clean layout that perfectly mimics a vintage invitation
  • The jack-o-lantern is quite dark, which may reduce its 'glowing' impact

Wan 2.5 (Preview)

  • + Dynamic lighting with impressive fire effects inside the pumpkin
  • + High level of detail in the background illustration, including the castle
  • + Rich, glossy textures on the central pumpkin
  • The main title text is cut off at the top border
  • Layout feels a bit cluttered compared to a professional invitation

Verdict: FLUX.2 [dev] is the clear winner as it functions perfectly as an invitation, with impeccable text rendering and a sophisticated, cohesive gothic design. While Wan 2.5 (Preview) has impressive illustrative details and lighting, its failure to keep the primary text within the frame significantly impacts its utility.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent source preservation overall
  • + Realistic hair texture that matches the lighting
  • + Creative interpretation of 'full, thick head of hair'
  • The afro hair texture doesn't naturally match the coarse, straight-leaning beard texture
  • The hairline integration at the temples is slightly blurry

Wan 2.5 (Preview)

  • + Natural and coherent hairstyle choice that matches the beard texture perfectly
  • + Exceptional integration of the hairline and sideburns with existing facial features
  • + Perfectly preserves original lighting and background
  • Slightly less 'thick' than the other model but arguably more realistic for the character

Verdict: Both models performed exceptionally well, perfectly preserving the source image's identity, lighting, and background. Wan 2.5 (Preview) is the winner because its hairstyle choice is much more cohesive with the subject's existing beard and facial features, whereas FLUX.2 [dev] added an afro that feels slightly disconnected from the beard's hair type.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Follows the isometric perspective and square diorama base request perfectly.
  • + Renders crisp, clean text and a well-proportioned flag icon.
  • + Materials look like realistic miniature PBR textures with a soft finish.
  • The composition feels slightly high in the frame, leaving more empty space at the bottom than the top.

Wan 2.5 (Preview)

  • + Distinct stylized 3D cartoon aesthetic with vibrant colors.
  • + High visual clarity and smooth lighting effects.
  • Failed the isometric requirement, opting for a standard eye-level perspective.
  • Text layout is cluttered with the flag icon squeezed next to the word SUSHI.
  • The background is a gradient instead of the requested solid light blue.

Verdict: FLUX.2 [dev] followed every technical instruction, including the specific isometric 45-degree angle, the square diorama base, and the solid background. Wan 2.5 (Preview) produced a high-quality stylized image but ignored the perspective and background constraints, resulting in a less accurate interpretation of 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 2.5 (Preview)
0% wins 0% ties 100% wins

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent caricature likeness of the original subjects face.
  • + Highly detailed and creative scene with the news desk on center ice.
  • + Perfectly integrates all requested themes: TV anchor desk, hockey rink/jerseys, and multiple dogs.
  • Text on the desk is garbled/unreadable.
  • The second human figure in the background is a bit distracting.

Wan 2.5 (Preview)

  • + Successfully incorporates the denim jacket from the source image into the caricature.
  • + Clean, vector-style artwork with bold colors.
  • + Includes all elements: anchor setup, hockey stick/TV, and dogs.
  • The facial likeness is significantly weaker and more generic than Model A.
  • The dogs look like generic clip art and lack the character visible in Model A.
  • The hockey integration is a bit more static, feeling like pasted-on props.

Verdict: FLUX.2 [dev] is the clear winner as it captures a genuine facial likeness of the woman in the source image while creatively blending the three themes into a single cohesive scene (a sports news desk on an ice rink). Wan 2.5 (Preview) produces a more generic 'cartoon' style that loses the specific features of the individual, although it did well to preserve the clothing choice from the original photo.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Natural and realistic fur texture across all animals
  • + Beautiful light interaction with backlighting and subtle god rays
  • + Excellent anatomical correctness for each animal type
  • The animals are mostly sitting rather than 'playfully chasing' or 'tumbling'
  • Added an extra bunny that wasn't explicitly requested

Wan 2.5 (Preview)

  • + Perfectly captures the 'playfully chasing' and 'tumbling' action from the prompt
  • + Vibrant colors and clear representation of all four requested species
  • + Very strong interpretation of 'god rays' and 'dew sparkles'
  • The fox kit has unnatural, glowing blue/orange eyes that look synthesized
  • Water droplets look like floating gems rather than natural dew
  • The kitten's anatomy and movement are slightly awkward

Verdict: Both models followed the prompt well, but FLUX.2 [dev] achieved a much higher level of photorealism with natural textures and lighting. While Wan 2.5 (Preview) better captured the dynamic action of chasing and tumbling, it suffered from 'AI-look' artifacts like glowing eyes and stylized floating droplets.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent adherence to the 'hand-painted' and 'pastel' art style requested.
  • + Perfectly captures the Ghibli-esque dreamy background with the rolling sky and flower field.
  • + Maintains the composition and distinctive poses of the original meme accurately.
  • The facial features are a bit generic and lack some of the expressiveness found in actual Ghibli character designs.
  • Replaced the urban street setting with a field, which may be a bit too much of a departure from the source.

Wan 2.5 (Preview)

  • + Strong character design that feels very close to modern anime/Ghibli protagonist styles.
  • + Preserves the original urban background (street, bus, people) much better than Model A.
  • + Great use of lighting and 'sparkle' effects to create a nostalgic mood.
  • The line work is a bit too clean and digital-looking compared to the requested 'hand-painted textures'.
  • Colors are slightly more saturated than the requested 'soft pastel' palette.

Verdict: Both models did an excellent job translating the 'distracted boyfriend' meme into an anime style while keeping the fundamental composition. FLUX.2 [dev] followed the stylistic prompt better by providing soft pastel colors and a hand-painted texture, though it completely changed the background. Wan 2.5 (Preview) produced much more expressive and accurate character designs that feel more 'Ghibli', while also maintaining the original urban context of the scene.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent depiction of wind-blown hair flowing symmetrically.
  • + High density of autumn leaves creates a clear sense of motion.
  • + Maintains very high color and texture consistency with the source image.
  • The transition where the hair meets the shoulders looks slightly digital and detached.

Wan 2.5 (Preview)

  • + Natural interpretation of wind direction with hair blowing primarily to one side.
  • + Preserves the subject's face and clothing with high fidelity.
  • + The added green leaves match the lush green background of the park better than autumn leaves.
  • The leaf rendering is quite sparse and some leaves look like simple green shapes.
  • The motion feel is slightly less 'energetic' compared to Model A.

Verdict: Both models followed the instructions well, maintaining the identity of the person and dog while successfully adding motion elements. FLUX.2 [dev] created a more dramatic, lively effect with a higher volume of flying leaves and more expansive hair movement, though the use of brown autumn leaves contrasts with the green trees in the background. Wan 2.5 (Preview) provided a more subtle and logically consistent edit with green leaves and a singular wind direction for the hair, but it lacks the dynamic impact of the former.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent typography including the requested accent on 'Caffè'
  • + Strong minimalist vector aesthetic with clean linework
  • + Accurately represents a classic cloche dome as a solid silhouette
  • The 'Est. 1720' text is slightly less balanced within its specific banner segment

Wan 2.5 (Preview)

  • + Good use of the 'warm brown and cream' color palette
  • + Pleasant paper texture effect on the background
  • + Legible and clean text rendering
  • The cloche dome is transparent, which is atypical for the traditional metal object requested
  • The banner construction is slightly awkward with the way the 'Est. 1720' segment overlaps

Verdict: FLUX.2 [dev] followed the prompt more effectively, producing a high-quality vector emblem with a solid cloche dome that matches the 'vintage minimalist' aesthetic. while Wan 2.5 (Preview) produced a nice color palette, its interpretation of the cloche as a glass dome was less conventional for this specific logo style.

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 2.5 (Preview)

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent vector iconography for individual stages
  • + Strong color palette adherence consistent with NASA themes
  • + Creative addition of the Apollo 11 crew names
  • Failed to follow the numbered sequence, displaying 9 confusing panels instead of 6
  • Major text hallucinations and illegible labels
  • Icon labels do not match the visuals correctly (e.g., Earth Orbit labeled under a non-Earth icon)

Wan 2.5 (Preview)

  • + Connected, logical infographic layout showing the progression of the mission
  • + Very clean, professional vector illustration style
  • + Mostly legible text and accurate stage labeling
  • The Saturn V rocket design is slightly distorted
  • Missed individual icons for 'Descent' and 'Landing', grouping them with text only
  • The crew portraits are significantly warped

Verdict: Wan 2.5 (Preview) produced a much more effective infographic by connecting the stages with a trajectory line, making the information flow logically. While FLUX.2 [dev] had higher quality individual icons, its layout was chaotic, the text was garbled, and it failed to follow the requested 6-step sequence accurately.

Next steps

Explore each model