Head to head
Esc

Models · slot A

to navigate to pick

FLUX.2 [dev] Flash fal Stable Diffusion 3.5 Medium Stability AI

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

FLUX.2 [dev] Flash

27.1 arena score

#5 of 62 in Text-to-Image

Skill signature · Text-to-Image

Stable Diffusion 3.5 Medium

16.8 arena score

#56 of 62 in Text-to-Image

Vote tally

Where the votes landed

FLUX.2 [dev] Flash

0%

win rate

Ties

0%

Stable Diffusion 3.5 Medium

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.2 [dev] Flash
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent photorealism with natural textures on the book, wood, and glass.
  • + Accurate physics, with the sphere resting naturally on the bottom of the cube.
  • + Perfect adherence to lighting and spatial instructions.
  • The plant is slightly less 'behind' the cube than 'beside' it, though still visible through the glass.

Stable Diffusion 3.5 Medium

  • + Good placement of the plant directly behind the cube as requested.
  • + Correct inclusion of all requested elements.
  • The blue sphere is floating unnaturally in the center of the cube.
  • Visual quality is lower, with a somewhat washed-out, digital look compared to the other model.
  • The perspective of the table edge is slightly warped.

Verdict: FLUX.2 [dev] Flash produced a significantly superior image with photorealistic textures and believable physics. While Stable Diffusion 3.5 Medium followed the placement of the plant more accurately, the floating sphere and lower overall resolution make it less compelling than FLUX.2 [dev] Flash's highly detailed and coherent render.

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 [dev] Flash
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent adherence to the 'repairing' prompt with detailed tools and hands-on interaction.
  • + Exceptional skin texture and facial realism for the elderly man.
  • + Perfectly balanced motion blur on background vehicles while maintaining a sharp subject.
  • The bike frame geometry is slightly distorted where the man is working.
  • Minor light artifacts around the background headlights.

Stable Diffusion 3.5 Medium

  • + Strong atmospheric mood with vibrant rain reflections on the pavement.
  • + Good sense of 'imperfect framing' and composition typical of candid street shots.
  • The man appears to be leaning on the bike rather than actively repairing it.
  • The character's hands and face lack detail and clear anatomy compared to the other model.
  • The bicycle wheels and frame have significant structural inconsistencies.

Verdict: FLUX.2 [dev] Flash captures the prompt with significantly higher technical accuracy, providing a realistic 50mm look with convincing skin textures and clear 'repairing' actions. Stable Diffusion 3.5 Medium achieves a nice cinematic mood through color and reflections, but fails on the detailed anatomy and the specific activity of repairing the bicycle.

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 [dev] Flash
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent photorealism in the skin texture and lifelike eyes.
  • + Very detailed engraving on the plate armor and visible fabric textures.
  • + Perfect adherence to the 'beads in hair' and 'torchlight' requirements.
  • The leather straps overlap in a way that looks slightly cluttered but remains coherent.

Stable Diffusion 3.5 Medium

  • + Strong, dramatic lighting and intense facial expression.
  • + High-contrast colors create a visually striking battle atmosphere.
  • Failed to include beads in the hair braids as requested.
  • The skin texture appears slightly more like paint or CG than a photograph.
  • The depth of field is less pronounced than requested.

Verdict: FLUX.2 [dev] Flash produced a significantly more detailed and prompt-accurate image, capturing specific elements like the hair beads and intricate leather textures with high fidelity. While Stable Diffusion 3.5 Medium has a strong artistic style and dramatic lighting, it missed several specific prompt descriptors and lacks the lifelike realism found in the FLUX.2 image.

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 [dev] Flash
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent typography rendering with readable sans-serif fonts
  • + High-quality, realistic food photography
  • + Clean professional layout that follows the requested sectioning
  • Confused content logic where all photos are pizza despite separate category headers
  • Small spelling errors in header text

Stable Diffusion 3.5 Medium

  • + Better variety of food items including salads and sides as requested
  • + Strong adherence to the grid-based layout concept
  • Poor typography and illegible text rendering
  • Low visual quality with smudged details and artifacts in the food photos
  • Layout feels cramped and lacks the 'modern minimalist' breathing room

Verdict: FLUX.2 [dev] Flash significantly outperforms Stable Diffusion 3.5 Medium in terms of graphic design quality and text legibility. While Stable Diffusion 3.5 Medium followed the prompt's request for food variety more accurately, FLUX.2 [dev] Flash produced a usable, professional-looking asset with superior image clarity and realistic lighting.

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 [dev] Flash
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent typography with a glowing, fiery text effect that matches the prompt perfectly.
  • + Clearly exploded burger view with suspended components as requested.
  • + Highly professional and balanced marketing layout.
  • The 'exploded' effect is slightly more static than dynamic, missing some of the 'motion' mentioned.

Stable Diffusion 3.5 Medium

  • + Strong fiery background with high contrast.
  • + Good quality texture on the burger bun and patty.
  • Failed the 'exploded' instruction; the burger is mostly assembled.
  • Text rendering is plain white and missing the requested fiery/glowing effect.
  • The price starburst is a simple vector style that clashes with the photographic image.

Verdict: FLUX.2 [dev] Flash significantly outperformed Stable Diffusion 3.5 Medium by following every detail of the prompt, including the specific 'exploded' layout and the fiery glowing text effects. Stable Diffusion 3.5 Medium failed to deconstruct the burger and ignored the stylistic requirements for the typography and the price starburst.

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 [dev] Flash
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent adherence to the requested text with zero spelling errors.
  • + Realistic chalk texture and handwriting styles that maintain consistency across the board.
  • + Great use of depth of field and environmental context for the café setting.
  • The date formatting is slightly different than the exact phrase 'APRIL 30, 2026' by using a dash instead of a colon.

Stable Diffusion 3.5 Medium

  • + Stylized decorative chalk art elements like borders and flourishes.
  • + High contrast between the chalkboard and the frame.
  • Major text rendering failures with significant spelling errors for almost every word.
  • Failed to follow the requested menu items correctly, inventing jibberish text.
  • The date and layout are cluttered and incoherent.

Verdict: FLUX.2 [dev] Flash followed the prompt meticulously, rendering all three complex menu items accurately and with realistic chalk textures. In contrast, Stable Diffusion 3.5 Medium failed significantly on text legibility and adherence, producing garbled words and incorrect pricing. FLUX.2's output is superior in both technical execution and prompt compliance.

The Reversed Rodeo

Text-to-Image

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

FLUX.2 [dev] Flash
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent texture rendering on the spacesuit and horse fur.
  • + High cinematic quality with a dynamic lunar surface background.
  • Significant anatomical error with a second horse tail appearing behind the rider.

Stable Diffusion 3.5 Medium

  • + Clean, minimalist composition that feels very surreal.
  • + Better technical performance regarding the single-subject horse anatomy.
  • Lower level of surface detail compared to the competitor.
  • The shadow of the horse on the clouds below feels physically inconsistent.

Verdict: FLUX.2 [dev] Flash offers a more detailed and cinematic environment, though it suffers from a major structural artifact with an extra horse tail. Stable Diffusion 3.5 Medium produces a cleaner subject but lacks the rich textures and atmospheric depth of the first image.

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 [dev] Flash
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent adherence to the passenger details, showing a bored woman looking at her phone.
  • + Very realistic lighting and textures for the capybara's fur and the leather steering wheel.
  • + Clearer taxi interior and exterior city lights that look authentic to a New York night scene.
  • The placement of the eyes and the snout on the capybara is slightly off-center compared to its body position.
  • The 'TX' sign on the roof appears inside the front window due to a confusing perspective.

Stable Diffusion 3.5 Medium

  • + Detailed rendering of the capybara's face and paws.
  • + Good use of depth of field to emphasize the driver.
  • Failed the prompt requirement for the passenger; she is looking forward rather than at her phone.
  • The capybara's paws are not clearly on the steering wheel, looking more tucked in.
  • The composition feels a bit cramped with the roof of the car being cut off.

Verdict: FLUX.2 [dev] Flash followed the complex prompt more accurately, specifically including the detail of the business woman looking at her phone, which adds to the intended humor of the scene. Stable Diffusion 3.5 Medium produced a high-quality subject but ignored the passenger's actions and had less convincing interior/exterior integration.

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 [dev] Flash
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent typography with nearly perfect spelling and formatting
  • + Atmospheric cinematic lighting that creates a professional, moody aesthetic
  • + Highly detailed and symmetric border incorporating thorns and webs as requested
  • Includes a small line of garbled text 'Burk: 0369' that was not in the prompt

Stable Diffusion 3.5 Medium

  • + Features more prominent twisted trees and a distinct parchment texture
  • + Vibrant colors and a classic illustrative style
  • Extremely poor text rendering with numerous spelling errors like 'Halloweeen Inviloween'
  • Incorrect date and location spelling ('The Aches' instead of 'The Arches')
  • Messy composition with overlapping elements

Verdict: FLUX.2 [dev] Flash significantly outperforms Stable Diffusion 3.5 Medium in this task, particularly in its ability to render complex text accurately and maintain a polished, cinematic composition. While Stable Diffusion 3.5 Medium captures the 'twisted trees' slightly better, its failure to spell basic words and the requested event details makes it unusable as an 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.2 [dev] Flash
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent text rendering and placement as requested.
  • + Superb material textures including the wooden base and rice grains.
  • + Perfectly followed the 45-degree isometric composition instruction.
  • The sushi roll construction is slightly surreal with fish wrapped around a roll.

Stable Diffusion 3.5 Medium

  • + Successfully rendered the word 'SUSHI' in a bold font.
  • + Met the requirement for a light blue background and white plate.
  • Missing the requested flag icon.
  • The text 'JAPAN' is very small and poorly aligned compared to the prompt.
  • Failed to include the requested wooden diorama base.

Verdict: FLUX.2 [dev] Flash followed every part of the prompt, including the specific text layout, flag icon, and the architectural diorama base. Stable Diffusion 3.5 Medium produced a simpler image that missed several key details like the wood base and the icon, while the text layout was less balanced.

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 [dev] Flash
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent adherence to the full animal list including rabbits and a tabby cat.
  • + Beautiful atmospheric lighting with clear god rays and dew sparkles.
  • + High anatomical accuracy for a multi-subject composition.
  • The scene is more of a static group portrait than an active 'tumbling and chasing' scene.
  • Duplicate animals (two foxes and two rabbits) were generated instead of one of each.

Stable Diffusion 3.5 Medium

  • + Vibrant color palette with a strong 'wholesome' vibe.
  • + Good fur texture on the puppy and fox.
  • Failed to include the rabbit and the tabby kitten is rendered as an orange fox-like kitten.
  • Heavy AI artifacts on the fox's legs and paws.
  • Overall higher level of saturation makes it look less 'hyper-photorealistic' and more like digital art.

Verdict: FLUX.2 [dev] Flash followed the complex prompt much more accurately, successfully including all requested species and capturing the specific lighting effects like dew sparkles and god rays. Stable Diffusion 3.5 Medium struggled with anatomical details and missed a key subject (the rabbit), resulting in a less realistic and more stylized image.

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 [dev] Flash
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent text rendering with accurate spelling of 'Caffè Florian' and 'Est. 1720'.
  • + Clean vector-style execution that matches the minimalist logo prompt.
  • + Accurate interpretation of the cloche dome with subtle steam.
  • The 'Est. 1720' text is placed below the banner rather than inside it as requested.

Stable Diffusion 3.5 Medium

  • + Ornate hand-drawn illustration style with good line weight.
  • + Sophisticated use of warm brown tones and cross-hatching for texture.
  • Multiple spelling errors including 'Florrian' and 'Est 170'.
  • The composition is cluttered and fails the 'minimalist' requirement.
  • Includes extraneous elements like a teacup not mentioned in the prompt.

Verdict: FLUX.2 [dev] Flash followed the instructions much better, producing a clean, professional logo with perfect spelling and a minimalist aesthetic. Stable Diffusion 3.5 Medium failed on critical details like text accuracy and simplicity, resulting in a cluttered design with typos.

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 [dev] Flash
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Flash

  • + Excellent text legibility and accuracy for most labels
  • + Accurate depiction of specific mission stages like the Saturn V launch and Lunar Module landing
  • + Color palette strictly follows the NASA-inspired prompt
  • Confusing layout with repetitive 'Lunar Orbit' and 'Descent' labels
  • Minor text corruption on 'Launch' and 'Moon'

Stable Diffusion 3.5 Medium

  • + Clean, minimalist vector aesthetic that feels modern
  • + Consistent iconography style across the left vertical axis
  • Poor typography with significant spelling errors and gibberish text
  • Icons do not accurately correspond to the assigned labels (e.g., Earth used for 'Landing')
  • Composition is unbalanced with a large empty white space

Verdict: FLUX.2 [dev] Flash is the clear winner as it successfully follows the requested infographic steps and produces mostly legible, accurate text related to the mission. While Stable Diffusion 3.5 Medium captures a nice minimalist style, its failure to render coherent text or relevant icons for the mission steps makes it an unsuccessful infographic.

Next steps

Explore each model