Head to head
Esc

Models · slot A

to navigate to pick

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

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

FLUX.1 [dev]

24.6 arena score

#16 of 62 in Text-to-Image

Skill signature · Text-to-Image

Stable Diffusion 3.5 Large Turbo

10.7 arena score

#61 of 62 in Text-to-Image

Vote tally

Where the votes landed

FLUX.1 [dev]

0%

win rate

Ties

0%

Stable Diffusion 3.5 Large Turbo

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 [dev]
Stable Diffusion 3.5 Large Turbo

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent photorealistic rendering of glass and light.
  • + Perfect adherence to spatial instructions with the book clearly on top of the cube.
  • + Highly realistic textures on the wooden table and red book cover.
  • The plant is behind the cube but doesn't show strong distortion/refraction through the glass itself.
  • The blue sphere appears slightly too large relative to the 'small' description.

Stable Diffusion 3.5 Large Turbo

  • + Clean, sharp aesthetic with high contrast.
  • + Good interpretation of the plant being visible through the glass pane.
  • Failed the spatial logic by putting the book and sphere inside the cube instead of the book on top.
  • Rendering looks more like a 3D digital model than a photograph.
  • The wooden table texture is repetitive and less natural.

Verdict: FLUX.1 [dev] followed every spatial instruction perfectly, creating a highly realistic scene with convincing lighting and material textures. Stable Diffusion 3.5 Large Turbo failed to place the red book on top of the cube, instead placing it inside with the sphere, and the overall image has a much more artificial, CGI-like 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.1 [dev]
Stable Diffusion 3.5 Large Turbo

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent adherence to the 'cinematic but realistic' requirement with natural lighting.
  • + Effective use of depth of field and motion blur to create depth and atmosphere.
  • + High-quality textures on the man's jacket and skin, feeling very lifelike.
  • The man is holding the handlebars rather than actively repairing a specific part of the bike.
  • Minor anatomy issues where the hands meet the handlebar grips.

Stable Diffusion 3.5 Large Turbo

  • + Successfully captured the 'repairing' action with a more hunched-over posture.
  • + The red color of the bicycle is vibrant and centered.
  • Has a plastic, 'cgi' look that contradicts the 'no stylization' and 'realistic' prompts.
  • Severe anatomical and structural errors, particularly with the bicycle frame merging into the man's leg.
  • Failed to create convincing 'motion blur from passing cars' or realistic rain effects.

Verdict: FLUX.1 [dev] produced a much more convincing and high-quality image that adheres to the photographic requirements of the prompt, featuring realistic rain and lighting. Stable Diffusion 3.5 Large Turbo struggled with the 'no stylization' constraint, resulting in a flat, artificial look with significant structural hallucinations involving the bicycle and the subject's legs.

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 [dev]
Stable Diffusion 3.5 Large Turbo

AI Judge Analysis

FLUX.1 [dev]

  • + Subtle, realistic skin textures including fine pores and freckles
  • + Masterful use of lighting and bokeh to create a cinematic atmosphere
  • + Exceptional eye detail and lifelike expression
  • Missed the request for beads in the hair
  • Armor looks more like general plate than specifically 'ornate engraved'

Stable Diffusion 3.5 Large Turbo

  • + Excellent adherence to the 'ornate engraved' plate armor request
  • + Clearly visible beads and ties in the hair
  • + Stronger 'battle-worn' feel with more prominent facial markings
  • Skin and hair have a slightly plastic, over-sharpened digital look
  • Lighting feels a bit flat compared to the requested warm torchlight
  • The ear and hair blending has some minor anatomical inconsistencies

Verdict: FLUX.1 [dev] produces a significantly more realistic and cinematically pleasing image with superior skin textures and lighting, though it fails to include specific details like hair beads. Stable Diffusion 3.5 Large Turbo adheres more strictly to every descriptive tag in the prompt, including the engraving and hair accessories, but the final image quality feels more like a 3D render than a lifelike photograph. FLUX.1 [dev] is the preferred choice for its sheer visual quality and atmosphere.

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 [dev]
Stable Diffusion 3.5 Large Turbo

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent typography and realistic layout that resembles a real menu.
  • + Clean white space and professional color palette.
  • + Includes clear headings for Appetizers and Mains as requested.
  • Most small text is illegible gibberish.
  • Missing the specific 'Pizza' section heading mentioned in the prompt.

Stable Diffusion 3.5 Large Turbo

  • + Better adherence to the 'grid' layout for food photos.
  • + Follows the section requirements including 'Pizza' and 'Mians' (Mains).
  • Composition feels cluttered and unbalanced with oversized food graphics.
  • Text rendering is poor with typos like 'Mians' and non-existent prices.
  • The 'grid' contains repeated, unrealistic-looking bowl assets.

Verdict: FLUX.1 [dev] produces a much more professional and realistic menu design that feels ready for commercial use, despite missing one section heading. Stable Diffusion 3.5 Large Turbo adheres more literally to the 'grid' and 'sections' request but fails on aesthetic quality with garish colors, cluttered layout, 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 [dev]
Stable Diffusion 3.5 Large Turbo

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent photorealistic rendering of food textures.
  • + Successful explosion/deconstruction effect with mid-air suspension.
  • + Accurately rendered most text elements requested including 'LIMITED TIME ONLY' and price.
  • Completely missed the 'MAGIC BURGER' title text.
  • The starburst for the price is missing, although the font and glow are decent.
  • The vertical stacking is a bit rigid and could be more dynamic.

Stable Diffusion 3.5 Large Turbo

  • + High energy fiery background with good ember effects.
  • + Vibrant colors and high contrast.
  • + Good sense of heat and lighting reflection on the burger.
  • Failed to include any of the requested text.
  • Failed to create an 'exploded' view; the burger is mostly intact though floating.
  • Illustrative style lacks the 'photorealistic detail' requested in the prompt.

Verdict: FLUX.1 [dev] followed the core layout instructions much better than Stable Diffusion 3.5 Large Turbo, providing the required 'exploded' view and a significant portion of the text. While FLUX.1 [dev] missed the primary title, its photorealistic execution of the food and price far surpasses Stable Diffusion 3.5 Large Turbo, which ignored all text requirements and the specific deconstructed composition.

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 [dev]
Stable Diffusion 3.5 Large Turbo

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent text rendering with accurate spelling of complex menu items.
  • + Authentic chalkboard texture and realistic handwriting variation.
  • + Strong adherence to the requested date and price formatting.
  • The 'cursive' request for the title is interpreted more as a stylized print than true elegant cursive.
  • Minor character artifacts in the bottom disclaimer text ('uufor' for 'us for').

Stable Diffusion 3.5 Large Turbo

  • + Aesthetically pleasing cafe-style composition with lighting and plants.
  • + Good chalk-like texture on the main headings.
  • Frequent spelling errors and nonsensical words like 'trulale' and 'ocotpg'.
  • Failed to include the specific year 2026 and accurate pricing.
  • Layout ignores the requested list structure in favor of columns with repetitive garbled text.

Verdict: FLUX.1 [dev] followed the prompt instructions with high precision, accurately rendering the specific menu items and date with legible, realistic handwriting. In contrast, Stable Diffusion 3.5 Large Turbo struggled significantly with text coherence, resulting in numerous spelling errors and a failure to complete the specific menu list requested.

The Reversed Rodeo

Text-to-Image

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

FLUX.1 [dev]
Stable Diffusion 3.5 Large Turbo

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent lighting and cinematic atmosphere with soft glowing edges
  • + Smooth anatomical integration between the horse and the astronaut's seat
  • + High level of detail in the mane and space suit textures
  • Failed the negative constraint: the astronaut is riding the horse, not the requested 'horse on top'

Stable Diffusion 3.5 Large Turbo

  • + Sharp, clear rendering of the space suit and horse's mane
  • + Creative use of shadows on the cloud layer below
  • + Included a planetary body in the background for composition
  • Failed the negative constraint: the astronaut is riding the horse, not the horse on top
  • Minor anatomical issues with the horse's front hoof and the astronaut's hand positioning

Verdict: Both models completely failed the specific spatial instruction for the horse to be 'on top' of the astronaut, instead providing the standard Interpretation of an astronaut on a horse. FLUX.1 [dev] is the slightly better image due to its superior lighting, cinematic depth, and more coherent anatomical details compared to Stable Diffusion 3.5 Large Turbo.

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 [dev]
Stable Diffusion 3.5 Large Turbo

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent photorealism in furry textures and skin tones
  • + Perfectly captures the 'bored' expression of the passenger
  • + High-quality rendering of the taxi atmosphere and bokeh city lights
  • The passenger appears to be in the front seat or mid-cabin due to perspective
  • The capybara's paws are not placed logically on the wheel

Stable Diffusion 3.5 Large Turbo

  • + Correctly places the human passenger in the backseat
  • + Anatomically better use of the capybara's paws on the steering wheel
  • + Stronger cinematic angle emphasizing the driving perspective
  • The passenger is very blurry and her bored expression/phone usage is less clear
  • The capybara's face is slightly less 'calm' and looks more like a 3D model than a real photograph

Verdict: FLUX.1 [dev] wins on pure image quality and expression, capturing the subtext of the prompt perfectly, although it struggles with the spatial depth of the car interior. Stable Diffusion 3.5 Large Turbo follows the spatial instructions of the passenger being in the back seat more accurately but fails to deliver the same level of photorealistic detail and clarity in the human character.

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 [dev]
Stable Diffusion 3.5 Large Turbo

AI Judge Analysis

FLUX.1 [dev]

  • + Successfully included all event details at the bottom.
  • + Captured a moody and cinematic lighting feel.
  • + Included the requested thorn border and twisted trees.
  • Several spelling errors in the main title and text.
  • Internal text repeats '7pm' twice.
  • The thorn border lacks the requested spider webs.

Stable Diffusion 3.5 Large Turbo

  • + Included the requested cobwebs in the border design.
  • + Clearer text rendering for the title section.
  • + Brighter, high-contrast composition.
  • Completely failed to include the scroll banner and event details (date, time, location).
  • The 'parchment' look is quite clean and lacks a vintage gothic feel.
  • Failed the prompt requirement for a gothic title by using a standard sans-serif font.

Verdict: FLUX.1 [dev] followed the prompt more closely in terms of content, successfully including all the specific event details and the scroll banner, despite some spelling issues. Stable Diffusion 3.5 Large Turbo completely ignored the latter half of the prompt regarding the specific text and event information, resulting in a generic image rather than a functional invitation.

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 [dev]
Stable Diffusion 3.5 Large Turbo

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent adherence to the 'soft refined textures' and 'miniature 3D cartoon' aesthetic.
  • + Accurately places the text and flag at the top-center as requested.
  • + Contains a realistic miniature diorama feel with high-quality PBR-like lighting.
  • Includes spelling errors in the secondary text ('SUSH CATON' instead of 'SUSHI').
  • The 'Japan' text is a bit small compared to the 'bold' requirement.

Stable Diffusion 3.5 Large Turbo

  • + Features very bold, clear 'JAPAN' text.
  • + Good 45-degree isometric composition.
  • + Clean, bright colors and sharp textures.
  • Fails to place the text at the top-center, integrating it into the scene instead.
  • Spells 'SUSHI' as 'SIIHI'.
  • The flag icon is a generic red and white rectangle rather than the Japanese flag.

Verdict: FLUX.1 [dev] followed the layout instructions much more accurately, placing the text and flag at the top-center rather than inside the diorama scene. While both models struggled with spelling the word 'SUSHI', FLUX.1 [dev] captured the requested soft, high-quality PBR miniature aesthetic and the Japanese flag more effectively than Stable Diffusion 3.5 Large Turbo.

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 [dev]
Stable Diffusion 3.5 Large Turbo

AI Judge Analysis

FLUX.1 [dev]

  • + Successfully includes all four requested animals: puppy, kitten, bunny, and fox.
  • + Captures the golden sunrise light and 'wholesome' atmosphere effectively.
  • + Good full-body compositions for all characters.
  • Has a very stylized, 3D-animation look rather than 'hyper-photorealistic'.
  • The animals have human-like paws and unnatural anatomy.
  • The kitten looks more like a second puppy.

Stable Diffusion 3.5 Large Turbo

  • + Significantly better fur texture and lighting on the animals.
  • + More realistic facial features and eyes compared to Model A.
  • Failed to include a bunny and a fox, showing only a puppy and two kittens.
  • Anatomical weirdness with overlapping limbs and paws.
  • The kitten on the right has hybrid ears that look like a caracal or lynx rather than a standard tabby.

Verdict: FLUX.1 [dev] followed the prompt more accurately by including all four distinct animals, whereas Stable Diffusion 3.5 Large Turbo missed half of the requested subjects. However, both models failed the 'hyper-photorealistic' requirement, with FLUX.1 [dev] opting for a Disney-style 3D render and Stable Diffusion 3.5 Large Turbo producing anatomical artifacts.

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 [dev]
Stable Diffusion 3.5 Large Turbo

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent minimalist vector style consistent with the prompt
  • + Accurate representation of an 'Est. 1720' banner and a cloche dome
  • + Correct tonal palette of warm brown and cream
  • Major spelling errors in the main text ('Caffé Flarilaan' and 'Reseaurant')
  • Addition of random numbers (11011, 1941) that were not in the prompt

Stable Diffusion 3.5 Large Turbo

  • + Strong text rendering with only a minor extra accent ('Caffeé Florian')
  • + Great use of subtle texture and vintage shading
  • + High-quality vector emblem illustration with good volume
  • The cloche dome is oddly integrated into what looks like a mug or cylinder
  • Text placement on the banner is slightly cramped

Verdict: While FLUX.1 [dev] nailed the minimalist layout and icon style perfectly, it suffered from severe hallucinations in the text rendering, misspelling both the brand name and the word 'Restaurant'. Stable Diffusion 3.5 Large Turbo followed the prompt with much better textual accuracy and a more sophisticated use of texture, despite a slightly confusing shape for the cloche dome.

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 [dev]
Stable Diffusion 3.5 Large Turbo

AI Judge Analysis

FLUX.1 [dev]

  • + Excellent adherence to the color palette with muted navy and reds.
  • + Clearer step-by-step layout structure with labeled steps.
  • + Good vector icon consistency and clean lines.
  • Text rendering is largely nonsensical despite an accurate header.
  • Includes irrelevant planetary icons like Saturn with rings.
  • The central rocket is a generic missile rather than a Saturn V.

Stable Diffusion 3.5 Large Turbo

  • + Stronger 'modern poster' aesthetic with bold, clear sections.
  • + Better thematic imagery including an astronaut on the lunar surface.
  • + Text sections attempt to categorize the mission stages more clearly.
  • Missed the step-by-step sequential instructions (only four steps shown).
  • Text rendering is garbled and includes typos like 'Apoll.o'.
  • The diagrammatic flow is less logical than requested.

Verdict: FLUX.1 [dev] followed the infographic structure much more effectively, providing a sequential path that matched the requested six steps, even if the icons were generic. Stable Diffusion 3.5 Large Turbo created a more visually striking poster with a better central illustration, but failed to include all the specific mission steps requested in the prompt.

Next steps

Explore each model