Head to head
Esc

Models · slot A

to navigate to pick

FLUX.2 [dev] Flash fal FLUX.2 [klein] 9B Black Forest Labs

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

FLUX.2 [klein] 9B

20.6 arena score

#13 of 32 in Image Editing

Vote tally

Where the votes landed

FLUX.2 [dev] Flash

0%

win rate

Ties

0%

FLUX.2 [klein] 9B

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] Flash
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent photographic realism with natural textures on the book and table.
  • + Perfectly adheres to all spatial and color prompts.
  • + The glass cube has realistic thickness and refraction.
  • The plant is slightly cut off at the top edge of the frame.

FLUX.2 [klein] 9B

  • + Strong composition and clear adherence to the color palette.
  • + Good handling of light and shadow on the wooden table.
  • The blue sphere appears to be floating unnaturally in the center of the cube.
  • The glass cube has rounded corners and thick walls that make it look more like a vase than a geometric cube.
  • The perspective of the book sitting on the cube is slightly tilted and awkward.

Verdict: FLUX.2 [dev] Flash is the clear winner as it produces a more photorealistic image where the objects interact naturally with physics, particularly the blue sphere resting on the bottom of the cube. FLUX.2 [klein] 9B fails to ground the sphere correctly, making it look like a floating digital artifact rather than a physical object.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent preservation of the man's identity, clothing, and hairstyle.
  • + Realistic interior integration with the man's hands on the wheel.
  • + Accurate representation of the California coastline atmosphere and lighting.
  • The car interior (wooden dashboard) does not match the modern interior of a Rolls-Royce Phantom Drophead Coupe.
  • Minor scale issue where the man appears slightly large for the cabin.

FLUX.2 [klein] 9B

  • + Perfectly preserves the exterior of the car from the source image.
  • + Captures the wide-angle motion of driving down a coastline excellently.
  • + Maintains the correct car model identity.
  • The man is barely visible, relegated to a small silhouette in the driver's seat.
  • Fails to showcase the man's specific features or clothing from the source image.

Verdict: FLUX.2 [dev] Flash is the clear winner for this edit as it successfully combines both source images, maintaining the specific identity and attire of the man while placing him in a believable driving scenario. While FLUX.2 [klein] 9B preserved the car's exterior better, it failed to meaningfully include the person from the second source image, making him an almost indistinguishable background element.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent depiction of motion blur from passing cars
  • + Natural and detailed skin texture on the face and hands
  • + Accurately captures the 'imperfect framing' request with a tight, slightly off-center crop
  • The anatomy of the bicycle is slightly nonsensical near the handlebars
  • The background reflections are a bit muddy compared to the pavement detail

FLUX.2 [klein] 9B

  • + Strong composition with a good use of shallow depth of field
  • + The bicycle has a more coherent structure
  • + Detailed ground reflections and textures that enhance the rainy atmosphere
  • Fails to include the requested motion blur from passing cars
  • The framing feels a bit too perfect and professional for the 'candid street photo' prompt

Verdict: FLUX.2 [dev] Flash captured the specific 'candid' and 'motion blur' requirements much better than the competitor, resulting in a more authentic street photography feel. FLUX.2 [klein] 9B produced a cleaner, more technically sound image of the bicycle itself, but it largely ignored the request for motion blur.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent realization of the leather strap and metal buckle textures
  • + Accurate depiction of hair beads and braiding as requested
  • + Deeply lifelike and expressive eyes
  • The scars appear slightly more like fresh blood/wounds than battle-worn scars
  • Skin texture is slightly softened in focus compared to the more rugged second image

FLUX.2 [klein] 9B

  • + Superior metal engraving detail and reflection work
  • + Skin texture and scarring feel more rugged and 'battle-worn'
  • + Excellent depth of field with cinematic lighting
  • Hair beads are very small and less distinct than requested
  • A minor logical artifact where the torch on the right seems to merge into the character's shoulder

Verdict: Both models performed exceptionally well on this complex prompt. FLUX.2 [dev] Flash delivered slightly better technical adherence to the jewelry and leather details, while FLUX.2 [klein] 9B produced a more atmospheric image with superior armor engraving and skin realism. FLUX.2 [klein] 9B is the winner for its more convincing 'battle-worn' aesthetic and the intricate beauty of the plate armor.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent use of vibrant secondary colors to differentiate blocks.
  • + Very clean grid layout with consistent image sizing.
  • + Modern typeface choice matches the 'bold sans-serif' prompt.
  • Nonsense header text 'CASSLAL 5 Menu'.
  • The prices and text underneath headings are largely illegible gibberish.

FLUX.2 [klein] 9B

  • + Stronger typography for the main headers.
  • + Better representation of 'minimalist' aesthetic through white space.
  • + Includes a variety of food items beyond just pizzas.
  • Duplicate 'Appetizers' heading in the grid section.
  • Layout feels slightly more cluttered with the long paragraph of text at the top.
  • Text rendering becomes very messy and broken in the bottom half.

Verdict: Both models struggled with legible body text, but FLUX.2 [dev] Flash delivered a more polished and professional design layout that better utilized vibrant accent colors as requested. FLUX.2 [klein] 9B followed the minimalist instruction well but suffered from redundant headings and an awkward distribution of photos in its grid.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent text integration with a realistic fiery glow effect
  • + Perfect 'exploded' view with clearly separated and suspended components
  • + High photorealistic detail on the textures of the meat and lettuce
  • The starburst design for the price is slightly less dynamic than the title text

FLUX.2 [klein] 9B

  • + Very dynamic composition with a strong sense of motion through radial streaks
  • + Bold, readable typography for the price starburst
  • + Vibrant colors and high contrast
  • Does not follow the 'exploded' burger request as the layers are mostly touching
  • Text for 'MAGIC BURGER' looks slightly more like a 3D bevel than actual fire

Verdict: FLUX.2 [dev] Flash is the clear winner for its superior adherence to the 'exploded' layout, providing a clean separation of ingredients that matches the cinematic ad style perfectly. While FLUX.2 [klein] 9B has great energy and color, it failed to separate the burger components as requested in the prompt, resulting in a more standard stacked burger profile.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent chalk texture with realistic smudging and dusting on the board.
  • + Perfect spelling and coherent formatting of all requested menu items.
  • + Natural integration into a café environment with depth of field and warm lighting.
  • The 'Today's Specials' heading is less elegant/cursive compared to the other model.

FLUX.2 [klein] 9B

  • + Beautiful cursive script for the title as requested in the prompt.
  • + Highly legible handwriting throughout the entire board.
  • + Captures the 'smudge' effect from board cleaning very effectively.
  • Contains a spelling error in the bottom line ('fress' instead of 'fresh').
  • The composition is a bit flat and clinical compared to a real café setting.

Verdict: Both models followed the prompt exceptionally well, producing realistic chalk handwriting with no digital font artifacts. FLUX.2 [dev] Flash is the winner because it maintained perfect spelling and felt more like a real photograph within a café, whereas FLUX.2 [klein] 9B had a spelling error ('fress') which detracts from the realism of a professional menu 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] Flash
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent character likeness, preserving the facial features and accessories perfectly
  • + Matches the studio lighting and background of the original images
  • Major composition failure where it overlays the character on top of the original person
  • Fails to replicate the specific pose from Image 1, creating a seated posture instead

FLUX.2 [klein] 9B

  • + Successfully replicates the complex dynamic pose from Image 1
  • + Accurately transfers clothing, scarf, and sunglasses from Image 2
  • + Seamlessly integrates the character into the lighting and environment
  • Minor facial likeness distortion compared to the source image
  • Anatomy issue with the right hand appearing as a clenched, simplified fist

Verdict: FLUX.2 [klein] 9B is the clear winner as it followed the core instruction to recreate the character in the specific dynamic pose of the reference image. FLUX.2 [dev] Flash failed the task by simply overlaying the character on top of the original woman, resulting in a nonsensical image with extra limbs and heads.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent cinematic lighting and a more cohesive, high-quality aesthetic.
  • + The composition feels more balanced with a clear sense of scale above the planet.
  • + Very clean rendering of the horse and the astronaut suit without distracting elements.
  • Failed the specific spatial instruction for the horse to be on top of the astronaut.
  • Followed a traditional interpretation of 'horse riding' rather than the surreal inversion requested.

FLUX.2 [klein] 9B

  • + High level of cosmic detail including multiple galaxies and nebulae.
  • + Sharp rendering of the astronaut's face visible through the visor.
  • Failed the specific spatial instruction for the horse to be on top of the astronaut.
  • Includes distracting, nonsensical gibberish text at the bottom of the frame.
  • Over-crowded background creates a busy and less cinematic composition.

Verdict: Both models failed the negative constraint to have the horse on top of the astronaut, instead opting for the common 'astronaut riding a horse' trope. FLUX.2 [dev] Flash won this comparison because of its superior cinematic quality, cleaner composition, and lack of the distracting hallucinations/text present in the FLUX.2 [klein] 9B output.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Successfully identifies the coat and scarf textures.
  • + Adds more complex jewelry as requested.
  • Severely distorts the subject's face with grotesque anatomical glitches.
  • Changes the subject's hair and adds a duplicate head structure behind the primary one.
  • Fails to maintain the background and original person's likeness.

FLUX.2 [klein] 9B

  • + Successfully preserves the subject's face, hair, and unique skin patterns almost perfectly.
  • + Great job matching the lighting on the subject to the environment.
  • + Correctly adapts the coat, scarf, and jeans from Image 2 to the pose in Image 1.
  • The jewelry is somewhat different in style compared to Image 2 (beaded vs metallic).
  • The hand in the pocket lacks some structural detail.

Verdict: FLUX.2 [dev] Flash completely fails the task by creating a horrific anatomical distortion of the subject's face and changing their identity. In contrast, FLUX.2 [klein] 9B performs an excellent edit, perfectly preserving the base person's likeness and skin details while accurately transferring the clothing from the second image.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent photorealism in the capybara's fur and the taxi interior
  • + Correctly places the passenger in the back seat as requested
  • + Cinematic lighting and high-quality bokeh in the background
  • The capybara's paws look more like human-animal hybrid claws than realistic capybara feet
  • The 'TX' sign on the roof is an unusual interpretation of taxi signage

FLUX.2 [klein] 9B

  • + The capybara paws look slightly more natural in their placement on the wheel
  • + High detail in the passenger's face and phone glow
  • Prompt adherence failure: the passenger is sitting in the front seat next to the driver instead of the back seat
  • The viewpoint makes it look like the camera is inside the engine or dashboard looking through the windshield

Verdict: FLUX.2 [dev] Flash followed the instructions much more accurately than FLUX.2 [klein] 9B, specifically placing the passenger in the back seat and providing a realistic interior view. While both models produced high-quality, photorealistic textures, FLUX.2 [dev] Flash managed the complex spatial arrangement of the prompt better, despite slightly uncanny rendering of the capybara's paws.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent typography style that fits the gothic theme
  • + Very detailed spiderweb and thorny border
  • + Good lighting on the jack-o-lantern
  • Includes some hallucinated gibberish text in the event details section
  • The parchment effect is less noticeable than the competitor

FLUX.2 [klein] 9B

  • + Perfect text rendering for all requested details with no errors
  • + Stronger parchment texture and vintage aesthetic
  • + Better composition with the inclusion of a full moon in the background
  • The thorny border is slightly less intricate than Image A
  • The shadows on the pumpkin are a bit softer, reducing some drama

Verdict: Both models followed the prompt exceptionally well, but FLUX.2 [klein] 9B is the winner because it successfully rendered every piece of text correctly, whereas FLUX.2 [dev] Flash included illegible characters in the date/time section. FLUX.2 [klein] 9B also provided a more complete 'vintage poster' feel with its parchment texture and atmospheric moon.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent preservation of the source background and clothing
  • + Successfully added a massive amount of hair as requested
  • The hairline is very harsh and lacks natural blending at the forehead
  • The texture of the hair feels slightly more like a wig or an 'afro' filter rather than a natural growth of the subject's hair type

FLUX.2 [klein] 9B

  • + Natural-looking hair texture that matches the beard's curl pattern
  • + The hairline and sideburns are realistically integrated with the existing facial features
  • + Excellent preservation of lighting and source details
  • One small stray hair strand at the top looks slightly disconnected

Verdict: FLUX.2 [klein] 9B provides a far more convincing edit by matching the texture of the hair to the subject's existing beard and creating a believable, soft hairline. While FLUX.2 [dev] Flash followed the instructions, the result looks like a digital overlay rather than a natural part of the person's head.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Perfectly rendered flag icon.
  • + Higher quality textures on the fish and rice grains.
  • + Excellent typography and layout.
  • The sushi roll construction is slightly nonsensical with fish draped over a cut roll.

FLUX.2 [klein] 9B

  • + Accurate variety of sushi types including maki and nigiri.
  • + Good 3D cartoon aesthetic with soft lighting.
  • + Clean isometric composition.
  • Incorrect flag icon (resembles Yemen or a generic tricolor instead of Japan).
  • Slightly muddy wood texture on the diorama base.
  • Maki rolls appear slightly flat or low-poly.

Verdict: FLUX.2 [dev] Flash is the superior model in this prompt, primarily due to its perfect rendering of the Japanese flag and significantly higher quality textures for the food and wood materials. While FLUX.2 [klein] 9B provides a good isometric layout, it fails on the specific requested detail of the flag icon and has lower overall visual clarity.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Expertly integrates the hockey theme by placing the news desk directly on an ice rink with players in the background.
  • + Maintains a very strong facial resemblance to the source image while applying the bobble-head caricature style.
  • + Includes clever details like dogs holding hockey sticks and wearing jerseys.
  • The hands are slightly undersized and less detailed compared to the rest of the illustration.

FLUX.2 [klein] 9B

  • + Captures the caricature aesthetic well with vibrant colors and expressive features.
  • + Effectively incorporates the TV anchor desk and the crowd atmosphere.
  • The hockey theme is significantly less prominent, with only a few sticks visible in the crowd.
  • The facial features, particularly the eyes and mouth, deviate further from the source image's likeness.
  • The text in the image is nonsensical gibberish.

Verdict: FLUX.2 [dev] Flash is the clear winner as it perfectly balanced all three requested themes (TV anchor, dogs, and hockey) into a cohesive and creative composition. While both models followed the caricature style, FLUX.2 [dev] Flash maintained a much better likeness to the woman in the source image and integrated the hockey elements more imaginatively.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent depiction of morning dew sparkles on the grass
  • + Captures all four requested animals plus duplicates for a fuller scene
  • + Soft, beautiful backlighting with clear god rays
  • The animals are largely static and posing rather than 'playfully chasing and tumbling'
  • The fox proportions are slightly stylized/cartoonish

FLUX.2 [klein] 9B

  • + Better captures the action of 'playfully chasing' with more dynamic poses
  • + Higher contrast and more vibrant color palette in the meadow
  • + Excellent variety in butterfly colors and placements
  • Completely missed the 'baby bunny' requirement from the prompt
  • The kitten has an anatomical anomaly where its arm seems to blend into the fox's side

Verdict: FLUX.2 [dev] Flash adhered better to the prompt by including the bunny, which FLUX.2 [klein] 9B entirely omitted. While FLUX.2 [klein] 9B captured the 'playful' action more effectively, its failure to include a key subject and the anatomical clipping between the kitten and fox makes FLUX.2 [dev] Flash the more reliable masterpiece.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Perfectly captures the expressions and composition of the original meme.
  • + Excellent application of soft, textured painting style similar to Ghibli concept art.
  • + Preserves the nuanced facial details and clothing patterns from the source.
  • The lighting in the background is a bit washed out.
  • A strange small creature figure in the background replaces the people in a non-cohesive way.

FLUX.2 [klein] 9B

  • + Strong watercolor aesthetic and clean line art.
  • + Creative background replacement with a Ghibli-esque countryside and village.
  • + Good color palette that feels warm and nostalgic.
  • Completely loses the original character expressions, making everyone look happy and losing the meme's context.
  • The character in the red dress has a slightly flattened perspective compared to the original.

Verdict: FLUX.2 [dev] Flash is the clear winner for its ability to transform the image into an illustration style while perfectly preserving the source material's recognizable expressions and humor. FLUX.2 [klein] 9B creates a pleasant illustration, but by making all characters look smiling and friendly, it fails to maintain the core 'distracted boyfriend' narrative of the original photo.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Strong prompt adherence with dramatic wind-blown hair effect.
  • + Excellent preservation of the facial features and original background from the source image.
  • The flying leaves look a bit like static overlays rather than being integrated into the scene.
  • Some leaves overlapping the clothing have odd transparency/blending issues.

FLUX.2 [klein] 9B

  • + Dynamic hair movement that looks organic and windswept.
  • + High quantity of leaves creates a strong sense of energetic motion throughout the frame.
  • The leaf textures are slightly blurry compared to the subject.
  • Minor distortion at the tips of the hair strands where they meet the background.

Verdict: Both models followed the instructions effectively, adding wind-blown hair and flying leaves while maintaining the core identity of the source image. FLUX.2 [dev] Flash produces a cleaner subject, but FLUX.2 [klein] 9B creates a slightly more energetic atmosphere with a better distribution of leaves across the depth of the image.

Vintage Cafe Logo

Text-to-Image

“Vintage minimalist restaurant logo for "Caffè Florian", retro cloche dome with steam and "Est. 1720" banner, classic typography, warm brown and cream tones, subtle texture on light background, vector emblem style.”

FLUX.2 [dev] Flash
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent typography with a hand-drawn vintage feel
  • + Beautiful subtle grain texture and shading
  • + Superior composition using the banner for the main brand name
  • The steam is a bit wispy compared to the heavy linework of the cloche

FLUX.2 [klein] 9B

  • + Features a clean vector emblem style with a circular frame
  • + High contrast and very legible text
  • The 'Est. 1720' is on the banner while the main name is outside, reversing the typical hierarchy for this style
  • Lower complexity in the illustration's shading and texture

Verdict: FLUX.2 [dev] Flash produced a much more authentic vintage logo with sophisticated typography and texture that perfectly matches the requested 'classic' aesthetic. While FLUX.2 [klein] 9B is clean and functional, its layout is less harmonic and the illustration lacks the artistic depth found in the first image.

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
FLUX.2 [klein] 9B

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent text rendering with correct spelling of mission names and phases.
  • + Followed the color palette and requested flat-vector iconography accurately.
  • + Maintains a structured, logical sequence of the mission steps.
  • The 'Launch' label has some minor artifacting/overlapping text character issues.
  • Includes redundant icons for the same labels (two Earth Orbits, two Descents).

FLUX.2 [klein] 9B

  • + Strong minimalist aesthetic with clean, chunky vector shapes.
  • + Accurate NASA-inspired color palette usage.
  • Poor spelling in almost every text element ('AOILLO', 'EARDHT', 'TRANSLURAL').
  • The sequence of icons is confusing and does not logically follow the requested 1-6 mission steps.
  • The rocket icon looks more like a generic toy rocket than a Saturn V.

Verdict: FLUX.2 [dev] Flash is the clear winner as it successfully rendered almost all text correctly and followed the sequential logic of the mission steps. FLUX.2 [klein] 9B failed significantly on the text rendering and provided a disjointed layout that did not adequately illustrate the specific steps requested.

Next steps

Explore each model