Head to head
Esc

Models · slot A

to navigate to pick

FLUX.1 Krea [dev] Black Forest Labs Stable Diffusion 3.5 Large Stability AI

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

Stable Diffusion 3.5 Large

22.9 arena score

#29 of 62 in Text-to-Image

Vote tally

Where the votes landed

FLUX.1 Krea [dev]

0%

win rate

Ties

0%

Stable Diffusion 3.5 Large

0%

win rate

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]
Stable Diffusion 3.5 Large

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Perfect adherence to the spatial prompt with the book on top and sphere inside.
  • + Highly realistic textures on the wood, book cover, and glass.
  • + Sophisticated lighting and shallow depth of field.
  • The plant is more 'behind' than 'partially visible through' the glass due to framing.

Stable Diffusion 3.5 Large

  • + Clear visibility of the plant through the glass panels.
  • + Clean, modern aesthetic with sharp details.
  • Failed the spatial relational prompt by placing the book inside the cube and the sphere on top of the book.
  • The lighting feels a bit harsh compared to the 'soft window light' request.

Verdict: FLUX.1 Krea [dev] followed the complex spatial instructions perfectly, accurately placing the sphere inside the cube and the book on top. Stable Diffusion 3.5 Large failed these relational prompts, reversing the order of the objects and placing the book inside the cube. FLUX.1 Krea [dev] also produced a much more convincing photographic quality with naturalistic lighting.

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]
Stable Diffusion 3.5 Large

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent depiction of motion blur in the background traffic
  • + Natural skin textures and realistic facial hair on the subject
  • + Strong adherence to the 'imperfect framing' and 50mm lens look
  • The subject is holding the bike rather than actively repairing it
  • Some lighting inconsistencies on the man's hair given the dim environment

Stable Diffusion 3.5 Large

  • + Higher level of detail in skin texture and weathered hands
  • + More active interaction with the bicycle suggesting a repair action
  • + Strong rain effects and pavement reflections
  • Anatomical issue with the man's left leg appearing to vanish into the pavement
  • Background traffic lacks the requested 'motion blur'
  • The bicycle geometry is slightly warped near the handlebars

Verdict: FLUX.1 Krea [dev] captured the cinematic atmosphere better with accurate motion blur and a photographic 50mm feel, though it missed the specific action of 'repairing'. Stable Diffusion 3.5 Large delivered more impressive skin textures and an active pose, but suffered from significant anatomical clipping and failed to include the requested 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.1 Krea [dev]
Stable Diffusion 3.5 Large

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent adherence to the 'beads in hair' prompt
  • + Superior skin texture with realistic pores and faint scarring
  • + Highly atmospheric lighting with clear bokeh sparks
  • The engraving on the armor is a bit flat compared to the other model
  • The background is quite dark and lacks environmental context beyond lighting

Stable Diffusion 3.5 Large

  • + Ornate engraving on the armor is incredibly intricate and deep
  • + Good depiction of multiple material layers including chainmail and cloth
  • + Strong composition with a sense of a larger army in the background
  • Failed to include the requested small beads in the braided hair
  • The lighting feels more like daylight with a fire nearby rather than warm torchlight
  • The skin texture is slightly smoother and less weathered than requested

Verdict: FLUX.1 Krea [dev] followed the specific details of the prompt much more closely, particularly regarding the beads in the hair and the specific 'warm torchlight' atmosphere. While Stable Diffusion 3.5 Large produced more impressive armor engravings, it missed a key detail and provided a brighter, less intimate lighting setup than the 'close portrait' prompt suggested.

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]
Stable Diffusion 3.5 Large

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent structure that mimics a real-world vertical menu layout.
  • + Food photography is well-integrated and professionally styled.
  • + Uses bold sans-serif fonts and vibrant orange accents as requested.
  • Text consists of gibberish characters upon closer inspection.
  • Includes 'Appetizers' as a section header twice.

Stable Diffusion 3.5 Large

  • + Features a very clean, high-end minimalist font for the main headers.
  • + The grid layout of food photos is vibrant and creates a strong frame.
  • + Includes specific sections for pizza as part of the visual grid.
  • The layout is more of a decorative poster than a functional, readable menu.
  • Significant spelling errors in headers like 'Appetizrs' and 'Maimaes'.

Verdict: FLUX.1 Krea (dev) produces a more practical and realistic menu layout that includes pricing and descriptions, even if the text itself is nonsensical. Stable Diffusion 3.5 Large creates a beautiful artistic composition with high-quality photography, but feels less like a functional menu due to its cramped central text column and spelling errors.

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]
Stable Diffusion 3.5 Large

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent text rendering and placement according to instructions.
  • + Strong composition that feels like a professional advertisement.
  • + Accurately represents the 'exploded' view with suspended components.
  • The '€6.99' is not inside a starburst as requested.
  • The fire effect feels a bit more like a static background layer than integrated lighting.

Stable Diffusion 3.5 Large

  • + High visual quality with realistic textures on the cheese and patty.
  • + Dynamic lighting with flames interacting with the burger components.
  • + Good sense of heat and motion with the embers and sparks.
  • Completely failed to include any of the requested text.
  • Failed to produce the 'exploded' view; the burger is mostly assembled.
  • The starburst element is missing entirely.

Verdict: FLUX.1 Krea is the clear winner as it successfully integrated all the complex text requirements and the specific 'exploded' layout requested. While Stable Diffusion 3.5 Large produced a visually stunning image with great textures, it completely ignored all typography and the structural requirement of suspended components.

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]
Stable Diffusion 3.5 Large

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent text legibility and accuracy of the specific menu items.
  • + Realistic chalk texture and handwriting variation as requested.
  • + Matches the specific date 'April 30, 2026' perfectly.
  • Includes several hallucinations of menu items not in the prompt.
  • The cursive for the title is quite basic rather than 'elegant'.

Stable Diffusion 3.5 Large

  • + Beautiful environmental composition showing the 'cozy café' context.
  • + Good visual balance and aesthetic appeal of the café setting.
  • Failed significantly on text rendering with many spelling errors (e.g., 'TODAAY', 'Ottpups', 'Cholulu').
  • Incorrect date (2024 instead of 2026).
  • The font appears more digital/uniform than natural chalk handwriting.

Verdict: FLUX.1 Krea [dev] followed the text instructions much more accurately, correctly rendering the specific date and most of the requested menu items with a realistic chalk texture. While Stable Diffusion 3.5 Large created a much more visually pleasing café environment, its inability to handle the complex text requirements of the prompt makes it less successful for this specific challenge.

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]
Stable Diffusion 3.5 Large

AI Judge Analysis

FLUX.1 Krea [dev]

  • + High visual clarity and realistic lighting on the space suit
  • + Followed the negative constraint partially by having the horse appear as a mechanical hybrid
  • + Excellent resolution and sharp textures
  • Failed the core 'horse on top' prompt instruction
  • Glitchy anatomy where the horse legs emerge from the mechanical torso

Stable Diffusion 3.5 Large

  • + Beautiful cinematic atmosphere with stardust and nebula effects
  • + Good composition with a sense of motion through space
  • + Coherent horse anatomy compared to the other model
  • Completely failed the 'horse on top' prompt instruction
  • Low contrast between the horse and the busy background

Verdict: Both models failed the specific spatial instruction 'horse on top, not vice versa,' instead providing the common interpretation of an astronaut riding a horse. FLUX.1 Krea (dev) has much better clarity and sharper details, whereas Stable Diffusion 3.5 Large has a more painterly, cluttered aesthetic that makes it harder to distinguish the subjects from the background.

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]
Stable Diffusion 3.5 Large

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Follows all prompt instructions, including the presence of the businesswoman in the back seat.
  • + The capybara's hands are correctly placed on the steering wheel.
  • + Captures the requested 'calm, professional expression' perfectly.
  • Included an extra woman in the front passenger seat which was not requested.
  • The taxi light on top of the car is oddly shaped and placed.

Stable Diffusion 3.5 Large

  • + Excellent fur texture and sharp lighting on the capybara.
  • + Colors are vibrant and the blurred background is well-executed.
  • Failed to include the businesswoman in the back seat entirely.
  • The paws are not correctly gripping the steering wheel as requested.
  • The capybara appears to have human legs and jeans, which looks uncanny.

Verdict: FLUX.1 Krea [dev] is the clear winner as it adhered to the complex prompt, successfully including the passenger in the back seat and the specific paw placement. Stable Diffusion 3.5 Large failed on several prompt requirements, most notably omitting the second subject and giving the capybara human legs.

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]
Stable Diffusion 3.5 Large

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent gothic typography that feels integrated into the design
  • + High-quality, polished illustration of the jack-o-lantern and bats
  • + Effective use of atmospheric lighting and a coherent spooky border
  • Several spelling errors and grammatical repetitions in the text
  • The main title is distorted into 'Pasty Halloween'

Stable Diffusion 3.5 Large

  • + Successfully renders the deckled edge parchment paper texture
  • + Detailed and moody background with the large moon and twisted trees
  • + The text in the scroll banner is perfectly legible
  • Completely failed to include the requested date, time, and location details
  • The central jack-o-lantern is off-center and small compared to the prompt's focus
  • Text font is a bit too modern and lacks a cohesive gothic aesthetic

Verdict: FLUX.1 Krea [dev] produces a much more balanced and professional-looking invitation layout with superior artistic polish despite the spelling errors. Stable Diffusion 3.5 Large captures the parchment texture well but misses several key text instructions from the prompt, including the specific event details and the central focus on the jack-o-lantern.

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]
Stable Diffusion 3.5 Large

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent adherence to the 'top-center large bold text' requirement.
  • + Perfectly captures the 'minimal' and 'clean' aesthetic requested in the prompt.
  • + Superior rendering of soft, refined PBR textures and gentle lighting.
  • The composition is very simple, featuring only one piece of sushi instead of a variety.
  • The flag is placed in the food rather than as a separate icon.

Stable Diffusion 3.5 Large

  • + Rich variety of sushi types and garnishes which feels more complete.
  • + Good use of perspective and includes more complex modeling of materials.
  • + Accurately includes more elements like chopsticks and a side bowl.
  • Failed to place text at the 'top-center'; instead, put it on a sign to the side.
  • The scene is cluttered and violates the 'minimal' requirement.
  • Background has a slight texture rather than being a 'solid' light blue.

Verdict: FLUX.1 Krea (dev) followed the layout and typography instructions much more accurately, yielding a clean, professional diorama that matches the 'minimal' and 'top-center' text requirements. Stable Diffusion 3.5 Large produced a more detailed and appetising food scene, but it failed on the specific text placement and overall minimalism requested in the prompt.

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]
Stable Diffusion 3.5 Large

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent texture on the fur, especially the fox and cats
  • + Strong lighting with clear god rays and golden hour atmosphere
  • + Very sharp and clean details on the animals and butterflies
  • Failed to include the baby bunny as requested
  • Generated two kittens instead of one kitten and one bunny

Stable Diffusion 3.5 Large

  • + Correctly included all four requested animals: puppy, kitten, bunny, and fox
  • + Highly dynamic composition with animals appearing to run toward the viewer
  • + Captured the 'dew sparkles' effect very effectively in the background
  • The fox's facial features are slightly distorted and look less realistic than the puppy
  • The kitten has a somewhat unusual facial structure compared to a real tabby

Verdict: Stable Diffusion 3.5 Large wins this comparison because it successfully included all four requested animals, whereas FLUX.1 Krea [dev] missed the bunny entirely. While FLUX.1 Krea [dev] produced slightly sharper fur textures, Stable Diffusion 3.5 Large better captured the specific lighting requests and the overall playful movement required by the prompt.

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]
Stable Diffusion 3.5 Large

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Excellent hand-drawn etching style that fits the vintage aesthetic.
  • + Beautifully detailed cloche with dynamic steam line work.
  • + Perfect application of subtle texture and color tones.
  • Serious spelling error in the primary text, spelling it 'FLANOR'IN' instead of 'Florian'.
  • The typography composition is a bit cluttered, overlapping the graphic element.

Stable Diffusion 3.5 Large

  • + Successfully spelled 'Caffé Florian' near perfectly.
  • + Clean, minimalist vector style that adheres well to a logo format.
  • + Accurate inclusion of all requested elements including the cloche, banner, and date.
  • The 'Caffé' has an extra 'e' appearing as 'Cafféé'.
  • The graphic design of the cloche and steam is very basic and lacks the 'vintage' charm of the competitor.

Verdict: Stable Diffusion 3.5 Large is the winner because it successfully followed the text instruction for the restaurant name, despite a minor double-vowel typo. While FLUX.1 Krea (dev) produced a far more beautiful and artistic vintage illustration, it failed significantly on the primary text prompt, resulting in gibberish.

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]
Stable Diffusion 3.5 Large

AI Judge Analysis

FLUX.1 Krea [dev]

  • + Features large, legible title text and numerical ordering
  • + Captures a playful, clean vector aesthetic
  • + Better adherence to the requested color palette
  • Icons do not match descriptions (e.g., Saturn with rings instead of Saturn V rocket)
  • Layout is disorganized with random numbering and nonsensical graphics
  • Lacks the technical 'infographic' feel requested

Stable Diffusion 3.5 Large

  • + Matches the layout and complexity of a technical infographic
  • + Includes a detailed rocket illustration and multiple planetary bodies
  • + Stronger visual composition with clear alignment and structure
  • Text consists entirely of garbled, illegible characters
  • Includes a space shuttle/generic rocket instead of the specific Saturn V requested
  • Fails to follow the specific 6-step sequence outlined in the prompt

Verdict: Both models failed significantly on the exact instructional content of the prompt, particularly regarding the specific steps and icons. FLUX.1 Krea produced a very simple design with legible numbers but irrelevant icons, while Stable Diffusion 3.5 Large created a visually complex and impressive technical poster that unfortunately contains purely gibberish text and ignores the step-by-step instructions. Stable Diffusion 3.5 Large is the likely winner for its superior composition and professional aesthetic, despite the lack of logical content.

Next steps

Explore each model