Black Forest Labs' open-weights image generation model with frontier performance, available for non-commercial local deployment
Settled by community votes across 19 shared challenges, with an AI judge weighing in on each.
FLUX.2 [dev]
#19 of 62 in Text-to-Image
Wan 2.6
#28 of 62 in Text-to-Image
Where the votes landed
FLUX.2 [dev]
0%
win rate
Ties
0%
Wan 2.6
0%
win rate
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.”
AI Judge Analysis
FLUX.2 [dev]
- + Excellent adherence to spatial prompts, placing the plant behind the cube.
- + High-quality glass rendering with realistic reflections and refractions.
- + Accurate interpretation of 'small blue sphere' relative to the cube size.
- − The plant is slightly less detailed in the background than the other model.
Wan 2.6
- + Beautiful natural lighting and shadows on the wooden surface.
- + Added character and realism to the book with worn edges.
- + Vibrant color saturation and sharp focus on the plant.
- − Failed the spatial prompt as the plant is placed to the side of the cube rather than behind it.
- − The 'small' blue sphere is relatively large, filling most of the cube's interior.
Verdict: FLUX.2 [dev] followed the spatial instructions perfectly, positioning the plant behind the glass cube to show transparency and keeping the sphere at a realistic 'small' scale. Wan 2.6 produced a very aesthetically pleasing image with great textures and lighting, but it missed the mark on the specific 'behind' positioning and the relative scale of the object inside the cube.
Man and Car in California
Editing“Make a photo of the man driving the car down the California coastline”
AI Judge Analysis
FLUX.2 [dev]
- + Excellent preservation of the man's hair texture, scarf, and coat pattern
- + Accurately recreates the interior details of the specific car model from the source
- + Great dynamic lighting on the subjects face and clothing
- − The man's facial features look significantly older/different compared to the source
- − The composition is very tight, showing less of the 'coastline' requested in the prompt
Wan 2.6
- + Perfectly captures the man's facial likeness and joyful expression from the source
- + Shows much more of the iconic California coastline with palm trees and winding roads
- + Maintains the car's exterior design and the man's specific clothing effectively
- − The man is seated on the passenger side (right-hand drive) which is incorrect for a California coastline setting
- − The motion blur on the front wheel is slightly inconsistent with the static background elements
Verdict: Both models successfully combined the two source images into a single scene. FLUX.2 [dev] excels at preserving the intricate textures of the clothing and car interior, but it fails to maintain the man's facial likeness. Wan 2.6 is the preferred choice because it perfectly preserves the man's face and captures the requested California coastline atmosphere much better, despite the curious choice to place him in a right-hand drive vehicle.
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.”
AI Judge Analysis
FLUX.2 [dev]
- + Excellent skin texture and hyper-realistic facial details
- + Accurate motion blur on the passing vehicle
- + Natural hand interaction with the bicycle components
- − The 'imperfect framing' request resulted in a very tight crop that cuts off the bicycle
- − The black jacket lacks the visible water droplets seen in the other model
Wan 2.6
- + Beautiful composition and atmospheric lighting with strong reflections
- + Great detail on water droplets clinging to the jacket and bike
- + Clearer storytelling with the subject crouching to work on the chain
- − Motion blur on the car looks static and slightly artificial
- − Face appears a bit smoothed compared to the natural skin texture of Model A
- − The tools on the ground look a bit distorted upon close inspection
Verdict: Both models captured the mood exceptionally well. FLUX.2 [dev] excels in raw realism, particularly with skin texture and the physics of the motion-blurred car, while Wan 2.6 provides a more cinematic and visually pleasing composition with better environmental interaction like rain droplets. FLUX.2 [dev] is the winner for its superior realism and adherence to the 'no stylization' instruction.
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.”
AI Judge Analysis
FLUX.2 [dev]
- + Excellent execution of complex engraving on the plate armor.
- + Beautifully rendered beads that look like actual jewelry woven into the hair.
- + Perfect facial symmetry and skin texture.
- − The 'faint scars' look a bit like clean makeup rather than battle wounds.
- − The leather straps look a bit too pristine and clean for a 'battle-worn' character.
Wan 2.6
- + Superb attention to the 'battle-worn' request with realistic soil and sweat on the skin.
- + Highly detailed fraying on the cloth underlayer and rough stitching on leather straps.
- + More dramatic and atmospheric use of warm torchlight and bokeh sparks.
- − The beads in the hair look slightly inconsistent in how they are attached.
- − Minor anatomical oddity on the side of the nose where lighting meets shadow.
Verdict: Both models performed exceptionally, but Wan 2.6 captures the 'battle-worn' aesthetic much clusters with grit, dirt, and frayed fabrics that feel authentic to the prompt. FLUX.2 [dev] produces a cleaner, more idealized 'high fantasy' look with superior hair-braiding and jewelry details, but lacks the gritty realism requested.
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.”
AI Judge Analysis
FLUX.2 [dev]
- + Organized layout and clear hierarchy.
- + Vibrant color accents that pop against the white background.
- + Includes all requested sections (Appetizers, Pizza, Mains).
- − Text rendering is largely gibberish.
- − Pricing values are unrealistic (e.g., $1300 for a starter).
Wan 2.6
- + Excellent professional design with a clean hero grid.
- + More realistic pricing and currency formatting.
- + High-quality, appetizing food photography.
- − Slightly cluttered header area with overlapping text.
- − Lacks a clear separate section for Pizza in the text area compared to Model A.
Verdict: Both models successfully interpreted the 'modern minimalist' prompt with vibrant accents and grid layouts. FLUX.2 [dev] followed the specific section names more accurately, but Wan 2.6 produced a far more professional and aesthetically pleasing design that looks like a real-world casual dining menu. Wan 2.6 is the preferred choice for its superior composition and realistic pricing details.
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.”
AI Judge Analysis
FLUX.2 [dev]
- + Excellent typography rendering with clean, glowing edges
- + High photorealistic detail on the burger patty and lettuce texture
- + Clean and professional composition suitable for a social media ad
- − The 'exploded' effect is very vertical and static compared to a dynamic action shot
- − The background is somewhat empty with less smoke and atmospheric depth
Wan 2.6
- + Dynamic and chaotic 'exploded' effect with tilted angles and flying ingredients
- + Rich atmospheric details with smoke, embers, and high-contrast fire
- + Creative integration of the fiery effect into the main title text
- − The pricing starburst looks a bit like a flat graphic compared to the 3D environment
- − The text 'LIMITED TIME ONLY' is placed very close to the bottom edge
Verdict: FLUX.2 [dev] produces a cleaner, more commercial-looking advertisement with superior font rendering and sharp details. However, Wan 2.6 captures the 'dynamic motion' and 'exploded' aspect of the prompt much better, creating an image with more energy and atmospheric depth. FLUX.2 is the likely winner for its overall professional polish and perfect adherence to the text requests.
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.”
AI Judge Analysis
FLUX.2 [dev]
- + Excellent chalk texture and smudge realism
- + Accurate spelling for all items including the cut-off prompt portion
- + Very clean and readable composition
- − The alignment of the second menu item is slightly awkward with the price floating isolated
- − The bottom sentence has a slightly more digital look compared to the top text
Wan 2.6
- + Beautiful cursive title that matches the elegant request
- + Excellent integration of the chalkboard into a physical café environment with wood framing
- + More natural-looking chalk dust and smudges at the bottom of the board
- − Contains minor spelling/rendering artifacts like the 's' in 'Specials' being slightly malformed
- − The price for the second item is touching the dash
Verdict: Both models followed the complex prompt exceptionally well, rendering all text accurately and in the correct chalk style. FLUX.2 [dev] produced slightly clearer text and better spacing, while Wan 2.6 excelled in artistic composition, framing, and decorative chalk cursive.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
AI Judge Analysis
FLUX.2 [dev]
- + Excellent photorealism in the astronaut suit and horse textures.
- + Clear, cinematic composition with a distinct planet and nebula.
- + Accurate rendering of the horse's anatomy and reins.
- − Failed the negative constraint; the astronaut is riding the horse instead of the horse riding the astronaut.
Wan 2.6
- + Vibrant, artistic color palette and lighting.
- + Detailed tack and equipment on the horse.
- + Dream-like, surreal quality to the background clouds and nebulae.
- − Failed the negative constraint; the astronaut is riding the horse instead of the horse riding the astronaut.
- − The horse's back leg is anatomically awkward compared to the front legs.
Verdict: Both FLUX.2 [dev] and Wan 2.6 failed the logic test of the prompt, which specifically requested the 'horse on top' of the astronaut. Instead, both models produced high-quality, generic 'astronaut riding a horse' images. FLUX.2 [dev] is slightly preferred for its more realistic lighting and cleaner composition, whereas Wan 2.6 feels more like a digital painting.
Outfit Transfer Challenge
Editing“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.”
AI Judge Analysis
FLUX.2 [dev]
- + Successfully captures the complex layering including the gold jewelry and the specific 'INNERSY' brand waistband detail.
- + Matches the specific scarf pattern and pea coat style very closely.
- − Failed to preserve the subjects face, replacing it with the face of the man from Image 2.
- − Changed the hair texture and added a beard that was not in the source image.
Wan 2.6
- + Excellent preservation of the source person's identity, face, and vitiligo patterns.
- + Correctly integrates the coat and scarf while maintaining the original background and lighting.
- − Missed several layer details such as the jewelry and the specific waistband shown in Image 2.
- − The sunglasses style is a generic black frame rather than the brown semi-rimless style from Image 2.
Verdict: This challenge highlights a trade-off between outfit accuracy and identity preservation. FLUX.2 [dev] did an incredible job replicating every detail of the clothing, including the jewelry and brand text, but failed the 'source preservation' requirement by merging the second person's face into the first. Wan 2.6 perfectly preserved the subjects face and skin details but simplified the outfit, missing the accessories.
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.”
AI Judge Analysis
FLUX.2 [dev]
- + Excellent texture on the capybara's fur and the jacket fabric.
- + The lighting inside the car feels very natural and integrated with the city environment.
- + The woman's 'bored' expression perfectly matches the prompt's narrative request.
- − The perspective of the steering wheel looks slightly flat compared to the capybara's paws.
- − The taxi cap is a bit simplistic with just a 'T' icon.
Wan 2.6
- + Beautiful atmospheric effects with rain on the windshield and vibrant bokeh in the background.
- + Highly detailed taxi cap that looks like an authentic uniform piece.
- + Good placement of the car's interior elements like the rearview mirror and seatbelt.
- − The woman's expression looks slightly more distressed or confused than 'bored'.
- − The capybara's paws appear a bit claw-like and less natural on the wheel.
Verdict: Both models followed the complex prompt very well, but FLUX.2 [dev] is the winner due to its superior rendering of textures and its more accurate interpretation of the human passenger's bored expression. While Wan 2.6 added a nice rain effect, FLUX.2 [dev] felt more coherent as a single photographic shot.
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.”
AI Judge Analysis
FLUX.2 [dev]
- + Excellent text legibility and clean font choices
- + Strong composition with a clear focus on the central jack-o-lantern
- + Highly accurate rendering of the thorn and web border
- − The parchment texture is limited to the very edges of the frame
- − The color palette is somewhat muted and less cinematic
Wan 2.6
- + Rich cinematic lighting with realistic glows and shadows
- + Highly detailed twisted trees and atmospheric depth
- + Beautiful gold texture on the title text
- − The 'You are invited' banner text placement is a bit cramped against the main title
- − The border is slightly messy with overlapping elements
- − Small details like the bats' faces are a bit distorted
Verdict: FLUX.2 [dev] excels in graphic design and typography, producing a very clean and professional-looking invitation that is easy to read. Wan 2.6 provides a much more atmospheric and cinematic illustration with impressive lighting and depth, though its text layout is slightly less polished than FLUX.2. Both models followed all complex instructions perfectly.
Bald man challenge
Image Editing“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”
AI Judge Analysis
FLUX.2 [dev]
- + Excellent preservation of the original facial features and expression
- + High-detail hair texture with realistic frizzy flyaways
- + Perfectly maintains lighting and background without any distortion
- − The Choice of an afro-style haircut feels slightly mismatched with the texture of the existing beard hair
Wan 2.6
- + Natural-looking hairstyle that complements the existing facial hair
- + Good volume and density as requested
- + Preserves the overall composition and background
- − Subtly alters the facial structure, making the man look younger and slightly different from the source image
- − The hair rendering looks slightly smoother/less detailed compared to the grit of the original photo
Verdict: Both models followed the instructions well, but FLUX.2 [dev] did a superior job of preserving the exact identity and facial details of the man from the source image. While Wan 2.6 provided a more conventional hairstyle, it slightly modified the man's face, whereas FLUX.2 [dev] successfully added hair as a perfect overlay to the original person.
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.”
AI Judge Analysis
FLUX.2 [dev]
- + Excellent 3D rendering with soft, refined textures
- + Perfectly followed text layout instructions including hierarchy
- + Superior composition with a more natural diorama feel
- − The sushi models are slightly more complex than a 'minimal' cartoon style might suggest
Wan 2.6
- + Strong isometric adherence and clean diorama base
- + Vivid colors and high clarity in the render
- + Good representation of a shrimp nigiri with cartoon styling
- − Failed the specific text hierarchy instructions by placing the flag next to 'Sushi' instead of below it
- − The text is very close to the top edge, affecting the balance
Verdict: FLUX.2 [dev] followed the prompt instructions much more accurately, particularly regarding the text layout and the specific requested hierarchy. While both models produced high-quality 3D isometric scenes, FLUX.2 [dev] had better centered composition and more refined PBR-style textures.
Over-the-top cartoon caricature
Editing“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.”
AI Judge Analysis
FLUX.2 [dev]
- + Excellent preservation of the subject's facial features in a caricature style
- + Highly creative integration of hockey by placing the news desk on an ice rink
- + Superior caricature technique with varied brushwork and classic 'big-head' proportions
- − Nonsense text on the front of the anchor desk
- − The extra character in the top left is slightly distracting
Wan 2.6
- + Strong, clean vector coloring and consistent art style
- + Good inclusion of professional equipment like the headset and microphone
- + Clear depiction of the hockey stick and puck
- − The facial resemblance to the source image is significantly weaker
- − The composition feels more like a generic cartoon than a personalized caricature
Verdict: FLUX.2 [dev] is the clear winner as it successfully maintains the recognizable facial features of the source woman while transforming her into a caricature. It also demonstrates more creativity by blending the news desk and hockey rink environments, whereas Wan 2.6 creates a more generic cartoon character that bears little resemblance to 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.”
AI Judge Analysis
FLUX.2 [dev]
- + Excellent detail in the fur texture and anatomy.
- + Subtle, realistic lighting that creates a natural atmosphere.
- + Clean composition with no major anatomical errors.
- − The animals are mostly sitting still, lacking the 'playfully chasing' and 'tumbling' action requested.
- − Included an extra five-legged appearance for the rabbit due to how they are huddled together.
Wan 2.6
- + Perfectly captures the 'playfully chasing' and 'tumbling' dynamic requested in the prompt.
- + Strong, dramatic god rays and sparkles that enhance the magical morning vibe.
- + Greater sense of energy and movement among the animals.
- − Some anatomical issues, such as the kitten's front paws being malformed.
- − The fox kit has an unnaturally long and thick body compared to its head.
- − Butterflies and petals in the air look a bit cluttered and messy in some spots.
Verdict: Both models followed the prompt well, but they succeeded in different areas. FLUX.2 [dev] produced a more technically clean and realistic image with better fur rendering, but failed to capture the action requested. Wan 2.6 accurately depicted the playfulness and motion of the prompt with much more dynamic lighting, making it the more representative interpretation despite some minor anatomical artifacts.
Studio Ghibli Anime Style
Editing“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”
AI Judge Analysis
FLUX.2 [dev]
- + Expertly captures the soft, atmospheric lighting and cloud style typical of Studio Ghibli.
- + Effectively replaces the urban background with a whimsical flower field that fits the dreamy mood.
- + Maintains excellent character poses and expressions from the source image.
- − The red dress was changed to a pinkish-coral color, losing some source fidelity.
- − The eyes of the central character are slightly less detailed compared to traditional Ghibli art.
Wan 2.6
- + Excellent hand-painted watercolor texture that feels authentic to concept art.
- + Preserves the original street setting and red dress color while still applying the requested style.
- + Superior character rendering that feels very close to high-end anime character design.
- − The white speckle/bokeh effect is a bit distracting and not strictly 'Ghibli-esque'.
- − The background characters are a bit simplified compared to the lead figures.
Verdict: Both models performed exceptionally well, capturing the essence of the 'distracted boyfriend' meme in a new art style. FLUX.2 [dev] leans more into the 'dreamy' aspect by replacing the background with clouds and flowers, whereas Wan 2.6 provides a beautiful watercolor texture that maintains better fidelity to the original source's colors and setting. Wan 2.6 is slightly preferred for its superior textural quality and preservation of the iconic red dress.
Golden Hour Stroll
Image Editing“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”
AI Judge Analysis
FLUX.2 [dev]
- + Successfully added a large volume of flying leaves into the scene
- + Significantly altered hair to show dramatic motion
- + Excellent source preservation of the background and subject features
- − The hair looks slightly stiff and unnatural in its outward spread
- − Most added leaves are brown/autumnal, which conflicts with the green summer trees shown
Wan 2.6
- + Natural interpretation of wind in the hair without over-distorting the face
- + Preserved the original source image perfectly while adding subtle motion elements
- + Consistent leaf colors matching the green park environment
- − The wind effect is very subtle compared to the 'energetic and lively' request
- − Fewer leaves added than requested
Verdict: FLUX.2 [dev] followed the instruction for an 'energetic' feel more effectively by adding a significant amount of hair movement and many flying leaves, though the brown leaf color choice was slightly odd for a green park. Wan 2.6 provided a more natural and photorealistic result, but it felt a bit too conservative given the prompt's request for dynamic motion. FLUX.2 [dev] is the winner for better adhering to the specific creative changes requested in the edit instruction.
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.”
AI Judge Analysis
FLUX.2 [dev]
- + Excellent layout with the banner integrated into the emblem style.
- + Perfect typography that matches the requested vintage aesthetic.
- + High-quality vector look with consistent subtle texture.
- − The steam icon is a bit small relative to the dome.
Wan 2.6
- + Elegant steam graphic on top of the cloche.
- + Good use of color gradients to suggest lighting on the dome surface.
- + Accurate text rendering for the main name.
- − The 'Est. 1720' banner is awkwardly tiny and disconnected from the main composition.
- − The background texture is a bit heavy-handed on the edges compared to a 'subtle' request.
Verdict: FLUX.2 [dev] followed the 'vector emblem' and 'banner' instructions much better, creating a cohesive and professional-looking logo. Wan 2.6 produced a decent illustration, but the layout was unbalanced due to the poorly placed and undersized banner element.
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.”
AI Judge Analysis
FLUX.2 [dev]
- + Excellent adherence to the complex multi-step instructions and iconography.
- + High-quality vector aesthetic with a clean, professional layout.
- + Accurate text rendering for the main titles and name tags.
- − The sequence of icons is scrambled and not in a logical grid order.
- − Some of the smaller labels contain gibberish text.
Wan 2.6
- + Successfully captured the requested color palette.
- + Readable text for the heads of the mission.
- − Failed to include any of the six requested infographic steps.
- − The image resembles a printed towel or fabric rather than a flat vector poster.
- − Weak composition with a large empty area in the center.
Verdict: FLUX.2 [dev] followed the detailed prompt requirements, producing multiple icons that correspond to the mission stages and maintaining a professional vector aesthetic. Wan 2.6 failed almost entirely on the prompt adherence, missing all the mission icons and producing an image that looks like a fabric print rather than a graphic design poster. FLUX.2 [dev] is the clear winner despite the disorganized layout of its icons.
Explore each model
Alibaba's multimodal generation model from the Wan AI suite, supporting text-to-video, image-to-video, reference-to-video with audio, and text-to-image, in both Chinese and English