Head to head
Esc

Models · slot A

to navigate to pick

FLUX.1 [schnell] FP8 Black Forest Labs FLUX.2 [klein] 4B Black Forest Labs

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

FLUX.1 [schnell] FP8

19.0 arena score

#48 of 62 in Text-to-Image

Skill signature · Text-to-Image

FLUX.2 [klein] 4B

22.3 arena score

#32 of 62 in Text-to-Image

Vote tally

Where the votes landed

FLUX.1 [schnell] FP8

0.0%

win rate

Ties

0.0%

FLUX.2 [klein] 4B

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.1 [schnell] FP8
FLUX.2 [klein] 4B

AI Judge Analysis

FLUX.1 [schnell] FP8

  • + Excellent handling of glass refractions and reflections
  • + Very high resolution and sharp textures
  • + Successfully placed the plant partially visible behind and through the glass
  • The glass object is more of a tall rectangular prism than a 'cube'
  • The inner shelf makes the composition more complex than a simple sphere in a cube

FLUX.2 [klein] 4B

  • + Correctly interpreted the 'cube' shape
  • + The composition is clean and follows the prompt's spatial instructions accurately
  • + Realistic texture on the red book cover
  • The plant is blurred in the background but not clearly visible 'through' the glass as requested
  • Lower overall resolution compared to Model A

Verdict: FLUX.1 [schnell] FP8 produces a much more striking image with superior lighting and material detail, though it struggles with the 'cube' geometry, rendering a tall rectangular container instead. FLUX.2 [klein] 4B followed the geometric instructions more literally but failed to realistically render the plant visible through the glass. FLUX.1 [schnell] FP8 is the preferred choice for its professional photographic quality and better handling of transparency.

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.1 [schnell] FP8
FLUX.2 [klein] 4B

AI Judge Analysis

FLUX.1 [schnell] FP8

  • + Excellent handling of wet pavement reflections
  • + Character skin texture feels natural and realistic
  • + High contrast cinematic lighting
  • Missed the request for motion blur on the passing cars
  • Anatomical issues where the hands join the handlebars
  • The bicycle geometry is slightly warped near the front wheel

FLUX.2 [klein] 4B

  • + Successfully captured the falling rain as requested
  • + Better posture representing someone actively inspecting/repairing a bike
  • + The bike design feels more authentic to a Japanese 'mamachari' style
  • Failed to include motion blur on cars
  • Background cars look static and a bit generic
  • The man's right hand holding the seat has anatomical errors

Verdict: Both models struggled with the specific request for motion blur, but FLUX.2 [klein] 4B followed the 'light rain' prompt more effectively by showing visible raindrops. FLUX.1 [schnell] FP8 produced a more punchy, cinematic image with superior reflections, but FLUX.2 [klein] 4B felt more like a candid street scene in inclement weather.

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.1 [schnell] FP8
FLUX.2 [klein] 4B

AI Judge Analysis

FLUX.1 [schnell] FP8

  • + Extremely high skin texture detail and realistic iris rendering.
  • + Dramatic, high-contrast lighting that perfectly matches the torchlight requirement.
  • + Exceptional rendering of fine details like facial hair and leather textures.
  • Hair lacks clear 'braids' with beads as requested in the prompt.
  • The composition is a bit too zoomed in, clipping most of the armor.

FLUX.2 [klein] 4B

  • + Excellent adherence to the braids and beads requirement.
  • + Shows much more of the 'ornate engraved plate armor' as requested.
  • + Effective inclusion of bokeh sparks and visible torchlight in the background.
  • Lighting is flatter and less cinematic than the competitor.
  • Facial features look slightly smoothed and less 'battle-worn' than the competitor.

Verdict: FLUX.1 [schnell] FP8 wins on raw visual fidelity and cinematic lighting, creating a very intense and lifelike character portrait. However, FLUX.2 [klein] 4B followed the specific structural instructions of the prompt much better, including the braids, beads, and a fuller view of the ornate 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.1 [schnell] FP8
FLUX.2 [klein] 4B

AI Judge Analysis

FLUX.1 [schnell] FP8

  • + Excellent adherence to the 'grid' layout for food photos.
  • + Includes all requested sections: appetizers, pizza, and mains.
  • + Very clean and professional minimalist aesthetic.
  • Several spelling errors in headers like 'APPTIZERS' and 'ORCETERS'.
  • The text density is a bit high, making small fonts hard to read.

FLUX.2 [klein] 4B

  • + High-quality food photography with vibrant colors.
  • + Dynamic layout that feels modern and high-end.
  • + Better rendering of specific bold sans-serif fonts.
  • Failed to include a section for 'Pizza' (used 'FRAZZA' and 'MALTS' instead).
  • Spelling in headers is significantly worse/more nonsensical ('MDAINS', 'MDANS').

Verdict: FLUX.1 [schnell] FP8 is the preferred choice because it strictly followed the structural requirements of the prompt, including all requested food categories (Appetizers, Pizza, Mains) in a coherent menu layout. While FLUX.2 [klein] 4B produced more visually striking food photography, it failed to categorize the items correctly and the text headers were less recognizable.

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.1 [schnell] FP8
FLUX.2 [klein] 4B

AI Judge Analysis

FLUX.1 [schnell] FP8

  • + Crisp and appetizing food textures
  • + Follows the circular composition common in food advertising
  • + Good flame assets in the background
  • Multiple spelling errors and redundant text lines
  • Incorrect price rendering and missed the starburst requirement for the main price
  • The 'exploded' effect is mostly just small crouton-like debris rather than main burger layers

FLUX.2 [klein] 4B

  • + Correctly renders the price text as '€6.99' within a starburst
  • + Captures the 'fiery, glowing effect' for the typography much better
  • + High motion energy with flying sparks
  • Major spelling errors in the primary title text
  • Lighting on the burger is a bit flat compared to the background intensity
  • The burger is not 'exploded' with components suspended separately

Verdict: Both models struggled with certain aspects of the prompt; FLUX.1 [schnell] FP8 produced a more realistic looking burger but failed significantly on the text requirements and price accuracy. FLUX.2 [klein] 4B followed the stylistic text requirements and price much better, despite the severe spelling errors in the main title. FLUX.2 [klein] 4B is the likely winner as its errors are common AI text artifacts while it hit more of the specific 'starburst' and 'glowing effect' creative requirements.

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.1 [schnell] FP8
FLUX.2 [klein] 4B

AI Judge Analysis

FLUX.1 [schnell] FP8

  • + Excellent chalk-like texture on the drawing surface
  • + Consistent font style throughout the menu items
  • + High clarity and resolution of the chalkboard frame
  • Poor spelling and duplicated text lines (e.g., 'Truffle mushroom ffle')
  • Text appears more like a clean digital whiteboard marker than actual chalk
  • Failed to render the title in the requested elegant cursive style

FLUX.2 [klein] 4B

  • + Successfully used elegant cursive handwriting for the text as requested
  • + Realistic chalk texture with smudges and dust on the board
  • + Much better adherence to the specific menu item names and prices
  • Minor spelling errors (e.g., 'Musheram', 'Ootpous', 'Brawn')
  • Spacing issues in the title line ('S PECIALS')

Verdict: FLUX.2 [klein] 4B is the clear winner because it followed the stylistic instructions for elegant cursive and realistic chalk texture. While it had some spelling errors, FLUX.1 [schnell] FP8's output was riddled with repetitive, garbled text and failed the cursive requirement completely.

The Reversed Rodeo

Text-to-Image

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

FLUX.1 [schnell] FP8
FLUX.2 [klein] 4B

AI Judge Analysis

FLUX.1 [schnell] FP8

  • + Successfully followed the challenging instruction of having the horse on top of the inanimate equipment/astronaut components.
  • + Captures a truly surreal and cinematic aesthetic with dramatic lighting.
  • + Shows a more creative interpretation of 'horse riding astronaut' by blending the two into a single unit.
  • The anatomy of the horses is somewhat distorted and abstract, especially the merging with the tech.
  • The 'astronaut' part is reduced to a mechanical block rather than a recognizable suit.

FLUX.2 [klein] 4B

  • + High level of anatomical detail on both the horse and the astronaut suit.
  • + Excellent background rendering of the galaxy and planetary curve.
  • Completely failed the negative constraint/specific instruction of 'horse on top, not vice versa'.
  • Provides a very literal and common interpretation rather than the requested surreal one.

Verdict: FLUX.1 [schnell] FP8 is the clear winner because it actually followed the specific and difficult spatial instruction to place the horse on top of the astronaut/equipment. While FLUX.2 [klein] 4B produced a cleaner and more traditionally beautiful image, it delivered a standard astronaut riding a horse, which was explicitly what the prompt asked to avoid.

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.1 [schnell] FP8
FLUX.2 [klein] 4B

AI Judge Analysis

FLUX.1 [schnell] FP8

  • + Features a clear 'TAXI' label on the hat as implied by the prompt
  • + Good color vibrancy and depth of field in the bokeh lights of Manhattan
  • The woman is holding two phones simultaneously, which is a major anatomical/logical error
  • The perspective makes the taxi appear much smaller than the passengers, as they take up the entire frame without a dashboard

FLUX.2 [klein] 4B

  • + Excellent anatomical execution with both paws correctly placed on the steering wheel
  • + Very realistic lighting and car interior geometry, including the center console and roof light
  • + Captures the 'bored' expression of the businesswoman perfectly
  • The cap is missing the specific 'taxi' wording, though it matches the style

Verdict: FLUX.2 [klein] 4B is the clear winner as it provides a much more coherent and realistic scene with correct passenger/driver placement. FLUX.1 [schnell] FP8 suffers from significant logic errors, such as the businesswoman holding two phones and a strange cropping that obscures the car's interior layout.

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.1 [schnell] FP8
FLUX.2 [klein] 4B
0% wins 0% ties 100% wins

AI Judge Analysis

FLUX.1 [schnell] FP8

  • + Features a vibrant glowing jack-o-lantern
  • + Composition feels balanced and neat
  • Significant spelling errors throughout the event details and banner
  • Border looks like brown fabric rather than thorny parchment
  • Background lacks detail and depth

FLUX.2 [klein] 4B

  • + Excellent atmospheric lighting and moody sky
  • + Intricate thorny and webbed border matches the prompt perfectly
  • + Superior vintage gothic aesthetic with detailed twisted trees
  • Primary title text contains several spelling errors
  • Minor typos in the banner and date

Verdict: FLUX.2 [klein] 4B captures the 'vintage gothic' and 'spooky but polished' aesthetic much more effectively with its atmospheric background and detailed thorny border. While both models struggled with some text accuracy, FLUX.1 [schnell] FP8 produced severe gibberish in the bottom half of the image, making the invitation unusable.

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.1 [schnell] FP8
FLUX.2 [klein] 4B

AI Judge Analysis

FLUX.1 [schnell] FP8

  • + Excellent miniature/isometric aesthetic with a raised diorama base
  • + High-quality soft 3D cartoon textures that match the prompt perfectly
  • + Clean composition with accurate lighting and shadows
  • The text is repetitive and misspelled ('JAPAN' and 'JAPAN' with a dot)
  • The second line of text is missing 'SUSHI' entirely

FLUX.2 [klein] 4B

  • + Successfully rendered the word 'JAPAN' in clear bold text
  • + Realistic materials on the plate and garnish (shiso leaves)
  • + Accurate 45-degree isometric perspective
  • The word 'SUSHI' is cut off to 'SUSH'
  • The flag icon is incorrect, appearing more like the flag of Austria than Japan
  • Lacks the 'miniature 3D cartoon' style requested, leaning more toward realistic CGI

Verdict: FLUX.1 [schnell] FP8 captured the requested '3D cartoon' and 'miniature diorama' style much better than its competitor, though it failed the text accuracy test. FLUX.2 [klein] 4B followed the text prompts slightly better (getting 'JAPAN' right) but failed to finish the word 'SUSHI' and provided an incorrect flag. Overall, FLUX.1 is preferred for its superior adherence to the stylistic and compositional instructions despite the typo.

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.1 [schnell] FP8
FLUX.2 [klein] 4B

AI Judge Analysis

FLUX.1 [schnell] FP8

  • + Excellent expressiveness in the animals' faces
  • + Warm, saturated color palette matches the 'golden sunrise' request perfectly
  • + Very high fur detail and texture consistency
  • Anatomical errors, including a strange fifth animal/hybrid head and a kitten with three paws in the foreground
  • Failed to include a bunny
  • The composition feels crowded and less photorealistic

FLUX.2 [klein] 4B

  • + Better movement and interaction between the animals
  • + Beautiful 'god rays' and lighting effects that feel more natural
  • + More distinct representation of the red fox kit
  • Failed to include the bunny
  • Includes two kittens instead of one kitten and one bunny
  • Fox ears and tail have some minor shape irregularities

Verdict: Both models failed to include the requested baby bunny, instead substituting extra kittens or unidentifiable creatures. FLUX.2 [klein] 4B is the better overall image as it manages a much more coherent composition with realistic lighting and avoids the severe anatomical glitches (like the floating head/extra limbs) found in the FLUX.1 [schnell] FP8 output.

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.1 [schnell] FP8
FLUX.2 [klein] 4B

AI Judge Analysis

FLUX.1 [schnell] FP8

  • + Successfully captured the requested architectural dome/cloche hybrid style
  • + Excellent warm brown and cream color palette
  • + Good vector emblem composition with banner
  • Significant spelling errors in the brand name ('Afe Flamilan')
  • Text overlaps and layout within the banner are messy

FLUX.2 [klein] 4B

  • + Very clean minimalist vector aesthetic
  • + Better background texture adherence
  • + Clearer rendered steam and cloche icons
  • Misspelled name as 'Caffè Flaxtion'
  • Redundant text with 'Est. 1720' appearing twice
  • Slightly less creative interpretation of the dome

Verdict: Both models struggled with the specific text 'Caffè Florian', with FLUX.1 [schnell] FP8 producing 'Afe Flamilan' and FLUX.2 [klein] 4B producing 'Caffè Flaxtion'. FLUX.2 [klein] 4B is the likely winner because its graphic elements are cleaner, the typography is more legible, and it better captures the requested vector minimalist style despite the redundant date text.

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.1 [schnell] FP8
FLUX.2 [klein] 4B

AI Judge Analysis

FLUX.1 [schnell] FP8

  • + Successfully uses the requested navy, white, and muted red color palette.
  • + Follows the horizontal linear progression of the infographic steps well.
  • + Captures the flat-vector, clean line aesthetic of modern infographics.
  • Text is largely gibberish except for headers.
  • Iconography is very abstract and doesn't clearly represent a Saturn V or Lunar Module in the timeline.

FLUX.2 [klein] 4B

  • + Closer representation of the specific objects like the rocket and Lunar Module.
  • + Includes labels for the crew members mentioned in the lore of the prompt.
  • + Generally readable text for title and some labels.
  • Significant spelling errors in the main title ('APLLO 1UT 1J').
  • Composition is cluttered and lacks a clear chronological flow for the steps.
  • Visual style is less 'flat vector' and more illustrative/painterly in sections.

Verdict: FLUX.1 [schnell] FP8 better adheres to the requested design aesthetic and color palette, creating a professional-looking layout despite the garbled text. FLUX.2 [klein] 4B provides more recognizable imagery and almost succeeds with text, but fails to maintain the infographic's logical flow and the specific 'flat-vector' style requested.

Next steps

Explore each model