Head to head
Esc

Models · slot A

to navigate to pick

FLUX.2 [klein] 9B Black Forest Labs Qwen Image 2512 Alibaba

Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.

FLUX.2 [klein] 9B

20.6 arena score

#13 of 32 in Image Editing

Skill signature · Image Editing

Qwen Image 2512

22.9 arena score

#30 of 62 in Text-to-Image

Vote tally

Where the votes landed

FLUX.2 [klein] 9B

0.0%

win rate

Ties

0.0%

Qwen Image 2512

100.0%

win rate

0.0% 0.0% ties 100.0%
Shared challenges 13

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 [klein] 9B
Qwen Image 2512

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Excellent handling of glass refraction and caustics on the wooden surface.
  • + Realistic soft lighting from the window side.
  • + The glass object looks like a high-quality physical vessel.
  • The sphere appears to be floating mid-air inside the glass, which looks slightly unnatural.
  • The 'cube' has rounded corners and thick walls, making it look more like a vase than a geometric cube.

Qwen Image 2512

  • + Perfect adherence to the 'cube' geometry with clean, sharp edges.
  • + The sphere is correctly resting on the bottom surface of the cube.
  • + Accurate spatial placement of all requested elements including the book and background plant.
  • The reflections inside the glass cube are a bit messy and inconsistent.
  • The lighting is slightly flatter compared to Model A.

Verdict: Both models followed the prompt perfectly in terms of object placement. FLUX.2 [klein] 9B produced a more photorealistic image with beautiful lighting and textures, but failed to make the glass a sharp cube. Qwen Image 2512 followed the geometric requirement of a 'cube' much better and placed the sphere realistically on the surface, making it the better choice for logical prompt adherence despite slightly lower render quality.

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 [klein] 9B
Qwen Image 2512

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Excellent skin texture and realistic hand details
  • + Successfully captures the action of repairing a bicycle
  • + Good execution of the shallow depth of field and rainy atmosphere
  • The motion blur on the passing cars is somewhat static compared to the sharp background buildings
  • The proximity of the man to the road edge feels slightly unnatural

Qwen Image 2512

  • + Natural street photography composition with an 'imperfect' feel
  • + Good reflection details on the wet pavement
  • Fails the main prompt instruction of 'repairing' as the man is just posing with the bike
  • Highly distorted hands and fingers
  • Missing the requested motion blur on the background cars

Verdict: FLUX.2 [klein] 9B followed the prompt significantly better by actually showing the subject repairing the bicycle with realistic manual dexterity and skin textures. Qwen Image 2512 struggled with anatomy, particularly the hands, and produced a static portrait that ignored the active verbs in the prompt.

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 [klein] 9B
Qwen Image 2512

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Excellent symmetrical composition and clear character focus.
  • + The engraving on the plate armor is crisp and highly detailed.
  • + Strong lighting contrast that clearly defines the character's facial features.
  • The scars look a bit like fresh red paint or superficial scratches rather than healed battle scars.
  • The background torches are somewhat distracting and lack depth.

Qwen Image 2512

  • + Atmospheric integration of sparks and torchlight creates a more moody, cinematic feel.
  • + Excellent texture work on the weather-worn leather straps and the grime on the skin.
  • + Braids and beads are more naturally integrated into the hairstyle.
  • The engraving on the armor is less sharp compared to the other model.
  • Slightly less symmetrical facial features, though this arguably adds to the realism.

Verdict: While FLUX.2 [klein] 9B provides a very clean and symmetrical portrait with incredibly sharp armor engravings, Qwen Image 2512 better captures the 'battle-worn' essence of the prompt through superior skin textures, more realistic scarring, and better atmospheric lighting. Qwen's interpretation feels more like a lived-in character, whereas FLUX.2 feels more like a polished digital painting.

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 [klein] 9B
Qwen Image 2512

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Successfully includes all section headers: Appetizers, Pizza, and Mains
  • + Clearer text rendering for the numbers and section titles
  • + Good use of vibrant accents and professional spacing
  • The food photos are repetitive, showing four pizzas instead of variety
  • Minor text hallucination in the main header

Qwen Image 2512

  • + Variety of food photos including appetizers, salads, and meats
  • + Grid layout for photos is well-organized and diverse
  • + Creative use of colorful shapes as accents beside menu items
  • Failed to include the specific 'Pizza' section header requested
  • Text rendering is very garbled and unintelligible compared to Model A
  • Menu item prices are nonsensical three-digit numbers

Verdict: FLUX.2 [klein] 9B followed the prompt's structural requirements much better by including all requested headers (Appetizers, Pizza, Mains) and providing much more legible typography. Qwen Image 2512 had better variety in its food photography grid and interesting layout elements, but it failed to follow the specific text instructions and the text quality was poor overall.

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 [klein] 9B
Qwen Image 2512

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Excellent text integration with consistent fiery glow effects across all elements.
  • + High-quality photorealistic texture on the meat patties and bun.
  • + Very clean composition with a strong sense of radial motion.
  • The 'exploded' effect is slightly less pronounced compared to Model B, with components still mostly stacked.
  • Includes double patties not specifically requested, though it adds to the 'magic' appeal.

Qwen Image 2512

  • + Successfully captures a more dramatic 'exploded' view with wider separation of elements.
  • + Excellent fiery starburst for the price point and detailed background environment.
  • + Very sharp text rendering with a distinct flame texture.
  • Missed the 'TIME' in the 'LIMITED TIME ONLY' text secondary message.
  • Some flying debris looks like ambiguous red blobs rather than food components.

Verdict: Both models followed the prompt well, producing high-quality commercial-style images. FLUX.2 [klein] 9B is the winner due to perfect text adherence and superior photorealism in the food textures, whereas Qwen Image 2512 failed to include one word of the required text and had slightly less appetizing rendering of the ingredients.

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 [klein] 9B
Qwen Image 2512
0% wins 0% ties 100% wins

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Excellent chalk texture with realistic smudges and erase marks on the board.
  • + Followed handwriting instructions with natural variations in letter size and slant.
  • + Accurate spelling throughout the menu including 'fresh' (though slightly stylized).
  • The cursive title 'TODAY'S SPECIALS' is a bit inconsistently weighted compared to the body text.

Qwen Image 2512

  • + Beautiful, consistent cursive handwriting for the body text.
  • + Well-composed layout and framing within a café environment.
  • + Perfect spelling and numerical rendering.
  • The text looks more like a digital font or a paint marker than actual dry chalk on a chalkboard.
  • The 'Risitto' misspelling (extra 'i') for Risotto.
  • Lacks the realistic chalk dust and smudge textures present in the other image.

Verdict: FLUX.2 [klein] 9B is the superior choice because it achieved the 'chalk texture' requested in the prompt, looking like a real smeared chalkboard, whereas Qwen Image 2512 rendered text that looks like a clean digital overlay. Additionally, FLUX.2 correctly spelled 'Risotto' while Qwen Image 2512 introduced a typo.

The Reversed Rodeo

Text-to-Image

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

FLUX.2 [klein] 9B
Qwen Image 2512

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Excellent background detail with numerous planets and nebulae creating a cinematic feel.
  • + Good rendering of the horse's anatomy and movement.
  • + Atmospheric lighting that matches the space environment.
  • Completely failed the negative constraint to put the horse on top of the astronaut.
  • Distracting gibberish text at the bottom of the image.

Qwen Image 2512

  • + High visual clarity and sharp rendering of the space suit and horse.
  • + Clean composition with a realistic-looking Earth horizon in the background.
  • Completely failed the specific prompt instruction to have the horse on top of the astronaut.
  • Anatomical error with the horse's front legs appearing poorly connected or distorted.

Verdict: Both models completely failed to follow the specific spatial instruction for the horse to be 'on top' of the astronaut, instead providing the conventional 'astronaut riding a horse' image. FLUX.2 [klein] 9B is slightly better due to its more creative and cinematic background, despite the unwanted text reaching for a movie poster aesthetic, whereas Qwen Image 2512 has noticeable anatomical issues with the horse's legs.

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 [klein] 9B
Qwen Image 2512

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Successfully places the passenger in the back seat as requested in the prompt.
  • + The lighting from the phone screen creates a realistic glow on the woman's face.
  • + Excellent texture on the capybara's fur and the fabric of the jacket.
  • The woman is sitting in the front passenger seat instead of the back seat.
  • The text on the taxi cap 'NEW TALA' is nonsensical.

Qwen Image 2512

  • + Captures a very 'bored' expression on the woman consistent with the prompt.
  • + High level of detail on the capybara's hands and the steering wheel texture.
  • + Strong cinematic lighting from the city street lights.
  • Failed to place the passenger in the back seat; she is in the front passenger seat.
  • The capybara's hands look somewhat like human fingers wrapped in fur or leather, which is slightly uncanny.
  • The perspective through the windshield makes the car look like it is stationary or facing the wrong way compared to traffic.

Verdict: Both models failed the spatial instruction to place the woman in the back seat, placing her in the front passenger seat instead. FLUX.2 [klein] 9B is the winner due to better overall composition, more realistic fur/hair rendering, and a superior interpretation of the lighting within the car cabin, whereas Qwen Image 2512 struggled with the anatomy 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 [klein] 9B
Qwen Image 2512

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Perfect text rendering for all lines including the date and location.
  • + High-quality composition with balanced light and shadow.
  • + Excellent adherence to the 'webs and thorns' border requirement.
  • The scrolling banner is a bit simple compared to the upper text.

Qwen Image 2512

  • + Elegant scroll banner design.
  • + Deep, atmospheric colors and moody lighting.
  • + Accurate event details at the bottom.
  • Significant spelling error in the main title: 'Hallowern'.
  • The thorns in the border look slightly less integrated with the parchment edge.

Verdict: FLUX.2 [klein] 9B is the superior image due to its perfect handling of the complex typography requested in the prompt, whereas Qwen Image 2512 failed on the primary title with a spelling error ('Hallowern'). Both models successfully captured the gothic aesthetic and specific background elements like twisted trees and glowing jack-o-lanterns.

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 [klein] 9B
Qwen Image 2512

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Excellent 3D wood texture and rendering
  • + Clean and professional typography layout
  • + Accurate 45-degree isometric projection
  • The flag icon is incorrect, resembling the Yemen flag instead of Japan
  • Slightly less variety in the sushi selection

Qwen Image 2512

  • + Correct Japanese flag icon
  • + Higher complexity in the miniature diorama with nice foliage details
  • + Stylized text matches the cartoon theme well
  • The diorama surface has a slightly 'melty' or inconsistent texture
  • Less refined PBR material feel compared to Model A

Verdict: Both models followed the prompt well, but Qwen Image 2512 is the winner because it successfully included the correct Japanese flag, whereas FLUX.2 generated an incorrect flag. While FLUX.2 had more 'realistic' clean wood textures, Qwen's interpretation of a 'miniature 3D cartoon scene' felt more complete with its added garnish and stylized text.

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 [klein] 9B
Qwen Image 2512

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Excellent dynamic composition showing the animals 'tumbling' as requested.
  • + Vibrant lighting with clear god rays and a joyful atmosphere.
  • + Large variety of colorful flowers and butterflies that fill the scene.
  • Missed the baby bunny character requested in the prompt.
  • The kitten's facial expression is slightly distorted/surprised compared to the others.

Qwen Image 2512

  • + Includes all four animals: puppy, kitten, fox, and bunny.
  • + Very high level of detail in the fur textures and eyes.
  • + Clearer, more realistic butterfly rendering on the left.
  • The animals are posed statically for a 'photo' rather than 'tumbling' and 'chasing' as requested.
  • The kitten's paws are merging awkwardly into the puppy's fur.

Verdict: Qwen Image 2512 followed the prompt's subject list more accurately by including the bunny, which FLUX.2 [klein] 9B missed. However, FLUX.2 [klein] 9B did a much better job capturing the requested action and vibe of the animals playing and tumbling, whereas Qwen Image 2512 opted for a staged portrait.

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 [klein] 9B
Qwen Image 2512

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Excellent adherence to the 'minimalist' and 'vector emblem' style keywords.
  • + Perfect text rendering with a clean, professional layout.
  • + Subtle, appropriate texture on the background.
  • The steam illustration is a bit too simplified and looks slightly detached from the cloche knob.

Qwen Image 2512

  • + Beautiful illustrative detail and vintage engraving style.
  • + Highly dynamic steam and sophisticated banner work.
  • + Rich, warm color palette that fits the 'vintage' prompt well.
  • Failed the 'minimalist' requirement, opting for a complex illustration over a simple logo.
  • The 'Est. 1720' text is slightly less clean than model A.

Verdict: FLUX.2 [klein] 9B is the winner because it strictly followed the 'minimalist' and 'vector emblem' instructions, resulting in a clean, functional logo. While Qwen Image 2512 produced a beautiful illustration with impressive detail, it was too complex for a minimalist logo request.

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 [klein] 9B
Qwen Image 2512

AI Judge Analysis

FLUX.2 [klein] 9B

  • + Stronger adherence to the flat-vector aesthetic with clean, bold shapes.
  • + Displays the requested color palette well with navy, muted red, and light gray.
  • + Includes the requested silhouette iconography for the crew members.
  • Significant spelling errors throughout the text (e.g., 'AOILLO', 'EARDHT', 'LANDINING').
  • The logical flow of the infographic follows a confusing path that is difficult to read.

Qwen Image 2512

  • + Layout follows a more logical sequential order for an infographic.
  • + Includes better details for the Lunar Module and Saturn V while maintaining a modern style.
  • + Generally better text rendering for the main title and most steps compared to the other model.
  • Included the instruction text 'Steps stop at landing' as literal text on the poster.
  • Step numbering is repetitive and slightly confusing (e.g., multiple 'Step 2' and 'Step 3').
  • The icons for the orbits are less consistent in style than requested.

Verdict: Qwen Image 2512 is the preferred choice because it maintains a logical structure and composition that works as an actual infographic, whereas FLUX.2 [klein] 9B has a very disorganized flow. While both models struggled with spelling, Qwen Image 2512 produced far more legible and accurate representations of the NASA equipment within a cohesive layout.

Next steps

Explore each model