Head to head
Esc

Models · slot A

to navigate to pick

FLUX.2 [dev] Turbo 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] Turbo

27.8 arena score

#4 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] Turbo

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] Turbo
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent adherence to spatial reasoning with the blue sphere resting naturally on the bottom of the cube.
  • + Superior lighting and texture, showing realistic reflections and dust on the glass surface.
  • + Very high visual fidelity in the wood grain and plant details.
  • The plant is slightly less 'behind' and more to the side than Image B.
  • Small optical distortion where the sphere meets the glass bottom is slightly unnatural.

Stable Diffusion 3.5 Medium

  • + Strong adherence to placing the plant directly behind the cube as requested.
  • + Colors are vibrant and follow the prompt requirements exactly.
  • The blue sphere is levitating mid-air inside the cube without any physical support, which looks surreal rather than realistic.
  • The glass cube has physics issues, where the top edge appears to merge into the book's cover.
  • Image is significantly lower in detail and resolution compared to the competitor.

Verdict: FLUX.2 [dev] Turbo produced a highly realistic, photorealistic image with excellent lighting and material textures, only slightly missing the 'directly behind' placement of the plant. Stable Diffusion 3.5 Medium followed the layout prompt more literally but suffered from major physics errors, such as a levitating sphere and distorted glass geometry. FLUX.2 is the clear winner for its superior visual quality and more believable scene composition.

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] Turbo
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent adherence to the 'motion blur from passing cars' prompt requirement.
  • + Incredible detail on the bicycle and the man's facial features and skin texture.
  • + The lighting and reflections on the pavement feel very realistic and ground the subject.
  • The bike's front wheel is clipping through the ground.
  • The man's kneeling position has some anatomical ambiguity where his legs meet the ground.

Stable Diffusion 3.5 Medium

  • + Beautiful cinematic lighting and color grading.
  • + Captures a strong sense of rainy atmosphere and shallow depth of field.
  • The man's hands are very poorly rendered and lack detail.
  • The bicycle geometry is warped and nonsensical compared to a real bike.
  • Lacks the requested 'motion blur' for the car in the background.

Verdict: FLUX.2 [dev] Turbo followed the prompt much more accurately, specifically regarding the motion blur of passing cars and the 'no stylization' request. While Stable Diffusion 3.5 Medium produced a moody, cinematic image, it failed significantly on the technical details of the hands and the bicycle structure.

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] Turbo
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Exquisite detailing on the engraved plate armor and leather straps.
  • + Perfectly captures the 'braided hair with small beads' requirement with distinct jewelry.
  • + Superior facial realism with lifelike skin texture, scars, and eyes.
  • The torch in the background is a bit generic in its flame rendering.

Stable Diffusion 3.5 Medium

  • + Strong highlights reflecting off the armor and face conveying a hot fire source.
  • + Intense, expressive eyes that fit the battle-worn theme.
  • + Good application of dirt and freckles on the face.
  • Missed the request for small beads in the hair braids.
  • Armor engraving is somewhat blurry and lacks the crispness of the competitor.
  • The bokeh sparks appear a bit clumpy and less like natural embers.

Verdict: FLUX.2 [dev] Turbo significantly outperformed Stable Diffusion 3.5 Medium by adhering to every specific detail of the prompt, most notably the beads in the hair and the high-fidelity leather and cloth textures. While Stable Diffusion 3.5 created a moody and intense portrait, it failed to render the beads and the plate armor engraving lacks the intricate clarity 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] Turbo
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent text legibility and graphic design hierarchy.
  • + High-quality, appetizing food photography that looks professional.
  • + Clean adherence to the grid layout and modern aesthetic.
  • One minor text overlap in the pink sub-header.
  • Prices are unusually high for a casual dining menu ($260 for a pizza).

Stable Diffusion 3.5 Medium

  • + Includes a large variety of food images.
  • + Follows the white background and minimalist prompt.
  • Text is largely unintelligible and includes many artifacts.
  • Image resolution is lower with a 'painterly' smearing effect on details.
  • Composition feels cluttered rather than professional.

Verdict: FLUX.2 [dev] Turbo significantly outperforms Stable Diffusion 3.5 Medium by producing a professional, usable menu design with clear typography and realistic photography. While Stable Diffusion 3.5 Medium provides a variety of images, the text is illegible and the overall layout lacks the polished graphic design quality found in FLUX.2 [dev] Turbo.

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] Turbo
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent adherence to the 'exploded' and 'suspended in mid-air' prompt requirements.
  • + High-quality rendering of fiery, glowing text as requested.
  • + Realistic food textures and dynamic sauce splashes.
  • The bottom bun appears slightly detached from the overall motion of the explosion.

Stable Diffusion 3.5 Medium

  • + Strong fiery background with vibrant colors.
  • + Clear text rendering.
  • Failed the 'exploded burger' requirement; the burger is mostly intact.
  • Text does not feature the requested fiery, glowing effect.
  • The starburst element is a simple graphic rather than a glowing integrated element.

Verdict: FLUX.2 [dev] Turbo followed every instruction, including the complex request for an exploded view and specific fiery text effects. Stable Diffusion 3.5 Medium produced a high-quality static image but failed to separate the burger components or apply the requested glow styles to the typography.

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] Turbo
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Perfect text rendering of the requested menu items and date.
  • + Highly realistic chalk texture with smudges and natural handwriting variations.
  • + Excellent adherence to the café background and depth of field.

Stable Diffusion 3.5 Medium

  • + Correctly identifies the chalkboard menu and wooden frame theme.
  • + Attempted a stylized cursive script for the headers.
  • Serious spelling errors and garbled text across the entire board.
  • The layout is cluttered and does not follow the requested three-item structure clearly.
  • Includes nonsensical symbols and digits that don't match the prompt.

Verdict: FLUX.2 [dev] Turbo followed the prompt with near-perfect accuracy, delivering legible text and a convincing chalk-on-board texture. In contrast, Stable Diffusion 3.5 Medium failed significantly on text rendering, producing various spelling errors and illegible characters.

The Reversed Rodeo

Text-to-Image

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

FLUX.2 [dev] Turbo
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent cinematic lighting and texture details on the spacesuit and horse fur.
  • + High resolution with a beautiful, complex space background.
  • + Dynamic and realistic composition.
  • Failed the negative constraint: the astronaut is on top of the horse, not vice versa.

Stable Diffusion 3.5 Medium

  • + Clean, high-contrast imagery.
  • + Good interpretation of the 'space' environment.
  • Failed the negative constraint: the astronaut is on top of the horse.
  • Significant anatomical errors with the horse's legs and the astronaut's lower body (extra leg visible).
  • Lower visual complexity compared to the other model.

Verdict: Both models failed the specific prompt instruction to place the horse on top of the astronaut, defaulting to the standard 'astronaut riding horse' trope. FLUX.2 [dev] Turbo is the clear winner as it produced a much higher quality, cinematic image with realistic textures, whereas Stable Diffusion 3.5 Medium suffered from major anatomical defects and a less polished aesthetic.

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] Turbo
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent adherence to the prompt, placing the passenger correctly in the back seat with a phone.
  • + Highly realistic texture for both the capybara's fur and the passenger's clothing.
  • + Very believable lighting and atmospheric integration of the Manhattan night background.
  • The paws on the steering wheel look slightly more like human hands than natural capybara feet.
  • The passenger is arguably in the front passenger seat rather than the back seat based on the proximity to the bird's-eye gear console and dash level.

Stable Diffusion 3.5 Medium

  • + Good character design for the capybara with a clear cap and jacket.
  • + Cinematic lighting with a warm glow on the capybara's face.
  • Failed to include the phone for the passenger as requested.
  • The passenger has an inquisitive/curious expression rather than a bored one.
  • The paws are not gripping the steering wheel in a realistic manner, appearing to just rest on top of a blurry ledge.

Verdict: FLUX.2 [dev] Turbo followed the prompt much more closely, successfully including the passenger's phone and bored expression while maintaining a high level of photorealism. Stable Diffusion 3.5 Medium missed several key details like the passenger's phone and struggled with the spatial orientation of the taxi's interior.

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] Turbo
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent typography with perfect spelling of all specific phrases.
  • + Highly detailed gothic assets including the thorn border and central jack-o-lantern.
  • + Atmospheric cinematic lighting that blends elements cohesively.
  • The parchment background is quite dark, which may reduce legibility for some.

Stable Diffusion 3.5 Medium

  • + Successfully captures the vintage parchment paper aesthetic.
  • + Included multiple jack-o-lanterns and twisted tree silhouettes.
  • Significant text rendering errors and misspellings throughout.
  • The composition feels less polished with erratic layout choices.
  • Failed to create a central jack-o-lantern as requested.

Verdict: FLUX.2 [dev] Turbo followed the prompt with high precision, delivering perfectly rendered text and a professional-grade composition. In contrast, Stable Diffusion 3.5 Medium struggled with the specific text requirements and the centered arrangement, resulting in a less coherent final image.

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] Turbo
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent typography with perfect spelling and layout.
  • + High-quality tactile textures that perfectly match the PBR material request.
  • + Includes all elements including the small flag icon and the raised diorama base.
  • None notable.

Stable Diffusion 3.5 Medium

  • + Features vibrant, glossy textures on the roe sushi.
  • + Clean solid background as requested.
  • Failed to include the Japanese flag icon.
  • The text 'SUSHI' has a graphical artifact on the 'S' and an extra dot on the 'I'.
  • Missing the requested 'raised diorama base' beneath the plate.

Verdict: FLUX.2 [dev] Turbo followed every instruction in the prompt perfectly, including the specific diorama base, flag icon, and complex typography. Stable Diffusion 3.5 Medium failed on several key details like the flag and base, and the text rendering was less precise than its competitor.

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] Turbo
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent adherence to the prompt by including all four requested animals correctly.
  • + Great rendition of lighting with clearly defined god rays and dew sparkles.
  • + Highly realistic fur textures and expressive, anatomically convincing eyes.
  • The butterfly wings look slightly generic compared to real monarchs.

Stable Diffusion 3.5 Medium

  • + Strong, vibrant color palette and warm lighting.
  • + Clean and simple composition suitable for a storybook.
  • Failed to include the requested baby bunny.
  • The cat looks more like a fox/cat hybrid with unusually large ears.
  • The fur texture appears more digital and 'brushed' rather than hyper-photorealistic.

Verdict: FLUX.2 [dev] Turbo significantly outperformed Stable Diffusion 3.5 Medium by following every detail of the prompt, specifically including all four animals (puppy, kitten, bunny, fox). FLUX.2 also achieved a much higher level of photorealism and captured the complex lighting effects like god rays and dew sparkles that the prompt requested, whereas Stable Diffusion 3.5 had a more simplified, illustrative look.

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] Turbo
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Perfect text rendering for all elements including the grave accent in 'Caffè'.
  • + Clean, professional vector-style composition that adheres strictly to the minimalist prompt.
  • + Excellent use of texture and warm tones to create a vintage feel.
  • The steam lines are a bit thick and stylized compared to the rest of the fine-line logo.

Stable Diffusion 3.5 Medium

  • + Intricate hand-drawn illustrative style with a classic feel.
  • + Good monochromatic color palette that fits the 'warm brown' request.
  • Significant spelling errors, including 'Florrian' and 'Est 170'.
  • The composition is cluttered and fails the 'minimalist' requirement of the prompt.
  • The cloche dome is distorted and includes unintelligible scribble text on the bottom banner.

Verdict: FLUX.2 [dev] Turbo followed every instruction perfectly, producing a professional-grade logo with flawless typography and a clean minimalist aesthetic. Stable Diffusion 3.5 Medium failed on most technical aspects, including spelling and prompt adherence for minimalism, resulting in a cluttered and inaccurate image.

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] Turbo
Stable Diffusion 3.5 Medium

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent text rendering with accurate labels for mission phases and crew.
  • + Successfully followed the request for specific icons like the Saturn V and Lunar Module.
  • + Clean, professional layout that functions logically as an infographic.
  • Includes a literal text string 'Saturn Vicon' due to a prompt misunderstanding.
  • The flow of the trajectory arcs is slightly cluttered and repetitive.

Stable Diffusion 3.5 Medium

  • + Followed the NASA-inspired color palette effectively.
  • + Maintains a consistent, stylized vector aesthetic across the icons.
  • Failed to render legible text or correct mission steps.
  • The icons do not accurately represent the Saturn V or the Lunar Module as requested.
  • Layout is confusing and does not clearly convey a temporal sequence.

Verdict: FLUX.2 [dev] Turbo produced a highly usable and accurate infographic with legible text and recognizable spacecraft icons, despite a minor literal interpretation of the prompt text. Stable Diffusion 3.5 Medium failed to follow the specific mission steps and produced illegible gibberish text throughout the image.

Next steps

Explore each model