Head to head
Esc

Models · slot A

to navigate to pick

FLUX.1 Krea [dev] Black Forest Labs FLUX.1 [schnell] FP8 Black Forest Labs

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

FLUX.1 Krea [dev]

18.8 arena score

#48 of 62 in Text-to-Image

Skill signature · Text-to-Image

FLUX.1 [schnell] FP8

19.0 arena score

#47 of 62 in Text-to-Image

Vote tally

Where the votes landed

FLUX.1 Krea [dev]

100.0%

win rate

Ties

0.0%

FLUX.1 [schnell] FP8

0.0%

win rate

100.0% 0.0% ties 0.0%
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 Krea [dev]
FLUX.1 [schnell] FP8

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent photo-realistic lighting and texture on the red book and wooden table.
  • + Accurate glass reflections including the sphere's caustic light on the base of the cube.
  • + Strong depth of field and soft lighting that matches the prompt perfectly.
  • The glass cube has a very thick bottom, making it look slightly more like a display case than a simple cube.

FLUX.1 [schnell] FP8

  • + Bright, clear colors and sharp focus throughout the scene.
  • + Effective placement of the plant behind the glass to show transparency.
  • The geometry of the 'cube' is tall and rectangular rather than a proper cube.
  • The internal shelf holding the sphere was not requested and adds unnecessary complexity to the composition.

Verdict: FLUX.1 Krea [dev] is the clear winner as it accurately depicted a cube and captured a much more realistic, high-quality aesthetic with convincing reflections. FLUX.1 [schnell] FP8 failed to generate a cube, instead creating a tall rectangular prism with an unprompted internal shelf.

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 Krea [dev]
FLUX.1 [schnell] FP8

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent depiction of rain with visible droplets hitting the subject.
  • + Stronger adherence to the 'motion blur from passing cars' instruction.
  • + The subject's facial features and clothing look more authentically Japanese and fit the requested 'elderly' description better.
  • The bicyle's handlebars are anatomically incorrect and merging strangely near the hands.
  • The man is standing with the bike rather than actively 'repairing' it.

FLUX.1 [schnell] FP8

  • + Natural skin textures and realistic lighting from brake lights on the wet pavement.
  • + Better bicycle geometry and handlebar structure.
  • + Good shallow depth of field effect.
  • Missed the 'motion blur' on the cars entirely; they appear static.
  • The man appears more Southeast Asian or South American than Japanese.
  • Fails to show actual rain falling, only the wet ground.

Verdict: FLUX.1 Krea [dev] captured the atmosphere of the prompt much better, successfully incorporating the rain particles and the motion blur of the traffic. While FLUX.1 [schnell] FP8 had better bicycle detail, it failed to execute the specific photographic techniques like motion blur and missed the ethnic descriptor of the subject.

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 Krea [dev]
FLUX.1 [schnell] FP8

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent composition that shows the full set of ornate engraved plate armor as requested.
  • + Captures the atmosphere perfectly with soft bokeh sparks and balanced torchlight.
  • + Provides very clear hair braiding with distinct beads as specified in the prompt.

FLUX.1 [schnell] FP8

  • + Intense and detailed eye rendering with strong micro-texture on the skin.
  • + Effective use of high-contrast warm lighting and shallow depth of field.
  • Fails to show the requested plate armor clearly, zooming in too close on the face.
  • Braided hair and beads are much less prominent than requested.
  • The lighting transition on the nose and forehead feels slightly artificial and overly sharp.

Verdict: FLUX.1 Krea [dev] is the clear winner as it followed all prompt instructions, including showing the ornate armor, leather straps, and specific hair details. FLUX.1 [schnell] FP8 zoomed in too far, losing the 'paladin' context and failing to emphasize the braids and beads as 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.”

FLUX.1 Krea [dev]
FLUX.1 [schnell] FP8

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent professional layout with a clean vertical grid of food photos.
  • + Strong use of bold sans-serif fonts and vibrant orange accents as requested.
  • + Realistic food photography that looks appetizing and professional.
  • Repeats the 'Appetizers' section heading twice instead of including 'Pizza'.
  • Text contains significant gibberish and spelling errors in the headers.

FLUX.1 [schnell] FP8

  • + Successfully includes sections for 'Apptizers', 'Pizzal', and 'Mains'.
  • + Good balance of text and images across a two-page spread style layout.
  • + Cleaner text rendering for prices and some headers.
  • Food photos are less realistic and look somewhat processed or artificial.
  • Included a strange 'SECCER' section heading which does not fit the prompt.
  • The layout feels a bit more cluttered despite being on a larger canvas.

Verdict: FLUX.1 Krea [dev] produces a much more realistic and professional-looking menu design that aligns better with the 'modern minimalist' aesthetic requested, despite a heading repetition error. FLUX.1 [schnell] FP8 followed the prompt regarding section variety more closely but failed on visual quality, featuring less appetizing food imagery and nonsensical headers like 'SECCER'.

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 Krea [dev]
FLUX.1 [schnell] FP8

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent typography with clean, readable, and mostly accurate spelling.
  • + Food photography is highly realistic and looks appetizing.
  • + Strong hierarchical composition that feels like a professional advertisement.
  • The price text failed to follow the 'starburst' instruction, placing it in a plain bubble instead.
  • The burger is only partially 'exploded,' with most layers still touching.

FLUX.1 [schnell] FP8

  • + Dynamic 'exploded' effect with many small flying ingredients conveying high motion.
  • + Includes a literal starburst for the price as requested.
  • + The burger lighting integrates well with the fiery background.
  • Significant text errors including 'LIIMITED' and 'LIMIED TIME NEEY'.
  • The price is confusingly rendered as both '€.99' and '€69' instead of '€6.99'.
  • The food rendering looks slightly more like a 3D model than a photorealistic burger.

Verdict: FLUX.1 Krea [dev] produced a professional-grade advertisement with clean typography and excellent food textures, let down only by a minor failure to put the price in a starburst. FLUX.1 [schnell] FP8 captured the sense of motion and the starburst element better, but failed significantly on spelling and price accuracy. FLUX.1 Krea [dev] is the winner for its superior visual quality and significantly more usable text.

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 Krea [dev]
FLUX.1 [schnell] FP8

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent chalk texture and realistic handwriting variation
  • + Clean layout with a convincing mahogany-style wooden frame
  • Included extra menu items not in the prompt
  • Minor spelling errors and gibberish in the smaller text lines

FLUX.1 [schnell] FP8

  • + Successfully captured the requested date and headers
  • + Natural lighting effects on the chalkboard surface
  • Significant spelling errors and text repetition throughout the menu
  • Font looks less like real chalk and more like a digital handwriting font
  • Layout is cluttered with overlapping and nonsensical pricing

Verdict: FLUX.1 Krea [dev] is the clear winner as it produces much more legible text and a sophisticated chalk aesthetic that feels authentic to a café. FLUX.1 [schnell] FP8 suffers from significant text hallucinations, repeating words and creating illogical pricing clusters.

The Reversed Rodeo

Text-to-Image

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

FLUX.1 Krea [dev]
FLUX.1 [schnell] FP8

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Strong cinematic lighting and high-quality textures on the horse and suit.
  • + Clear, sharp details against a realistic Earth backdrop.
  • + Successfully placed the horse and astronaut in space together.
  • Failed the negative constraint; the astronaut is riding the horse instead of the horse riding the astronaut.
  • Anatomical issues with the horse's legs merging into the equipment.

FLUX.1 [schnell] FP8

  • + Attempts a more surreal composition with a horse being carried on equipment.
  • + Good use of color and atmospheric lighting.
  • Failed the prompt; there is no astronaut visible, only two horses and mechanical parts.
  • Anatomical distortions where the horses merge with the equipment.
  • The second horse's neck and head appear disconnected from a body.

Verdict: Both models failed the complex spatial instruction to have a horse riding an astronaut. FLUX.1 Krea [dev] produced a much higher quality image with better textures and a clear astronaut, despite defaulting to the traditional 'astronaut riding horse' trope. FLUX.1 [schnell] FP8 failed to include an astronaut entirely, resulting in a confusing and anatomically mangled image of multiple horses.

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 Krea [dev]
FLUX.1 [schnell] FP8

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent photorealistic lighting and textures
  • + Perfectly captures the 'bored' expression of the passenger
  • + Accurate depiction of a yellow taxi hat and professional attire
  • Generated an extra passenger in the middle seat
  • The capybara's head shape is slightly distorted to look more human-like

FLUX.1 [schnell] FP8

  • + Strong composition that places the capybara and passenger in clear focus
  • + High detail in the capybara's fur and the taxi interior
  • + Accurate depiction of the Manhattan street through the window
  • The passenger is holding two separate phones, which is a logic error
  • The capybara's paws are not correctly placed on the steering wheel, with one floating/misaligned

Verdict: FLUX.1 Krea (dev) followed the aesthetic and mood requirements much better, creating a truly cinematic and photorealistic atmosphere. While it added an extra character, FLUX.1 [schnell] FP8 suffered from more significant technical artifacts, such as the passenger holding two phones and the capybara's misaligned paws on the wheel.

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 Krea [dev]
FLUX.1 [schnell] FP8

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent decorative border with intricate web and thorn details.
  • + Higher quality illustration style for the Jack-o-lantern and trees.
  • + Better adherence to the 'vintage gothic' aesthetic with parchment textures.
  • Several typos in the text including 'Pasty Halloween' and 'a nights Tright'.
  • The layout of the event details on the bottom scroll is cluttered and repetitive.

FLUX.1 [schnell] FP8

  • + Clean layout that feels more like a structured invitation.
  • + Good rendering of the moody, foggy atmosphere in the background.
  • Significant spelling errors throughout almost every line of text.
  • The border appears more like simple cloth folds rather than the requested webs and thorns.
  • The central pumpkin lighting is less 'cinematic' and looks a bit flatter.

Verdict: Both models struggled significantly with the text rendering, though FLUX.1 Krea [dev] (Model A) produced much higher quality artwork and better adhered to the decorative elements of the prompt like the thorns and webs. FLUX.1 [schnell] FP8 (Model B) followed the layout constraints but failed on the specific border requirements and generated highly garbled text for the event details.

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 Krea [dev]
FLUX.1 [schnell] FP8

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent text rendering with 'JAPAN' and 'SUSHI' correctly spelled and positioned.
  • + Accurate adherence to the small flag icon requirement.
  • + High-quality material textures on the salmon and rice.
  • The salmon texture looks slightly more realistic than 'cartoon' compared to the base.

FLUX.1 [schnell] FP8

  • + Strong isometric 3d cartoon aesthetic with soft, rounded forms.
  • + Pleasing layout of the sushi arrangement on the diorama.
  • Failed the text requirements, repeating 'JAPAN' with errors instead of 'SUSHI'.
  • Missing the explicit flag icon, instead attempting to integrate it into the text poorly.
  • The rendering of the sushi toppings is slightly plasticky and less refined.

Verdict: FLUX.1 Krea [dev] followed the prompt instructions perfectly, specifically excelling at the text rendering and the inclusion of the flag icon. FLUX.1 [schnell] FP8 captured the 'cartoon' style well but failed on almost all text and icon requirements, resulting in a less accurate output.

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 Krea [dev]
FLUX.1 [schnell] FP8

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Successfully includes the red fox kit and a clear golden retriever puppy.
  • + Better anatomical correctness for the kitten species.
  • + Excellent lighting with visible god rays and backlighting on fur.
  • Missed the baby bunny entirely, providing two kittens instead.
  • The jumping kitten in the background has a slightly unnatural pose.

FLUX.1 [schnell] FP8

  • + More expressive and playful facial expressions on the animals.
  • + Includes multiple species, though they are somewhat blended or indistinct.
  • + Rich, vibrant colors and a magical atmosphere.
  • Highly inaccurate species representation; the animals look like strange hybrids rather than a distinct retriever, kitten, bunny, and fox.
  • Failed to include a rabbit/bunny or a clear tabby kitten.
  • The animal in the bottom right has an anatomically distorted face.

Verdict: FLUX.1 Krea [dev] is the clear winner as it provides much higher photorealism and better species accuracy, even though it missed the bunny. FLUX.1 [schnell] FP8 produced a chaotic huddle of unrecognizable hybrid animals that looked more like AI-generated mashups than the specific creatures requested.

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 Krea [dev]
FLUX.1 [schnell] FP8

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent adherence to the 'vintage texture' request with a distressed paper effect
  • + High-quality hand-drawn illustration style for the cloche and steam
  • + Strong composition with a clear 'Est. 1720' banner as requested
  • The main text 'Caffè Florian' is misspelled as 'Café FLANDR'IN'

FLUX.1 [schnell] FP8

  • + Cleaner vector-like aesthetics that fit the 'logo' requirement
  • + Includes all requested elements like the cloche, steam, and banner
  • + Closer to the requested spelling, though still contains errors
  • The text 'AFE FLAMILAN' is poorly rendered and messy
  • The 'Est. 1720' is not inside the banner as specified by the prompt
  • Composition feels slightly unbalanced with repetitive decorative swooshes

Verdict: Both models failed to spell the name 'Caffè Florian' correctly, but FLUX.1 Krea [dev] (Image A) is the superior artistic choice due to its beautiful woodcut-style illustration and authentic vintage texture. While FLUX.1 [schnell] FP8 (Image B) has a cleaner 'vector' feel, its typography is disorganized and it failed to place the date on the banner as requested.

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 Krea [dev]
FLUX.1 [schnell] FP8
100% wins 0% ties 0% wins

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Strong aesthetic appeal with the large lunar surface foreground
  • + Clean 'Apollo 11' main header and legible numbering
  • + High-quality vector styling for the individual icons
  • Failed to provide the specific required icons like Saturn V
  • Nonsensical layout that jumps between icons without clear logical flow
  • Incorrect counting sequence (1, 2, 3, 4, 0, 6)

FLUX.1 [schnell] FP8

  • + Successfully followed the infographic structure with distinct steps and labels
  • + Excellent adherence to the NASA-inspired color palette
  • + Includes specific requested elements like Saturn V (text) and the lunar module
  • Text inside the infographic labels is largely garbled/gibberish
  • The layout feels a bit cramped at the bottom
  • The actual imagery for Earth orbit and Descent is abstracted and less crisp

Verdict: FLUX.1 [schnell] FP8 is the clear winner for its superior adherence to the 'infographic' prompt, capturing the sequence of events and information layout requested. While FLUX.1 Krea [dev] produces a more artistic and visually clean image, it fails significantly on logical flow, numbering, and icon accuracy.

Next steps

Explore each model