Black Forest Labs' state-of-the-art image generation model with maximum quality and speed, supporting text-to-image and multi-reference image editing with up to 4MP output
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX.2 [pro]
26.3
arena score
#8 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.2 [pro]
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.2 [pro]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [pro]
- + Perfect adherence to all spatial instructions
- + Excellent photographic realism and depth of field
- + Natural light and shadow interaction
- − None notable
Stable Diffusion 3.5 Large Turbo
- + Vibrant colors and clean lines
- + High resolution rendering
- − Failed spatial placement of the book
- − Failed spatial placement of the plant
- − The sphere appears to be inside the book rather than just inside the cube
Verdict: FLUX.2 [pro] followed the complex spatial instructions perfectly, placing every object exactly where it was requested with high realism. In contrast, Stable Diffusion 3.5 Large Turbo failed nearly all positional prompts, placing the book inside the cube and the plant to the side instead of behind it.
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 [pro]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [pro]
- + Excellent adherence to the 'cinematic but realistic' requirement with natural skin textures and lighting.
- + Correctly implements motion blur from passing cars and a shallow depth of field.
- + The composition feels like a genuine candid street photo with believable environmental details.
- − The main character's arm anatomy is slightly awkward where it meets the bicycle.
Stable Diffusion 3.5 Large Turbo
- + Successfully includes the red bicycle and the elderly man.
- − Fails to produce a realistic or 'no stylization' image, looking more like a 3D render.
- − Anatomy is severely broken, with the man's head blending into a hat-like shape and hands merging into the frame.
- − The environment lacks the requested motion blur, reflections, and cinematic atmosphere.
Verdict: FLUX.2 [pro] followed every detail of the prompt, including the complex technical requirements for motion blur, shallow depth of field, and a candid aesthetic. Stable Diffusion 3.5 Large Turbo failed significantly on realism and anatomy, providing a highly stylized, almost plastic-looking output that ignored the environmental lighting and texture requests.
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 [pro]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [pro]
- + Excellent adherence to lighting and atmosphere prompts like warm torchlight and bokeh sparks
- + High realism in facial textures, skin imperfections, and braided hair details
- + Exceptional surface rendering on the ornate engraved plate armor and worn leather straps
- − The bokeh sparks are slightly sharp for a shallow depth of field
Stable Diffusion 3.5 Large Turbo
- + Ornate armor design is visually interesting
- + Clear braided hair texture
- − Skin texture appears plastic and smoothed, missing the requested 'lifelike eyes' and realistic skin details
- − The red marks on the face look like flat face paint rather than 'faint scars and dirt'
- − Missing the 'warm torchlight reflection' and 'bokeh sparks' requested in the prompt
Verdict: FLUX.2 [pro] followed the prompt's atmospheric and textural requirements much more closely, resulting in a realistic, cinematic portrait with rich detail in the armor and skin. In contrast, Stable Diffusion 3.5 Large Turbo produced a stylized, plastic-looking image that missed several key descriptors such as the torchlight reflections and sparks.
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 [pro]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [pro]
- + Excellent adherence to the menu layout requested
- + Clear, readable sans-serif typography
- + Logical alignment of sections and food photos
- − Includes some spelling errors like 'Mins' and 'MageFiza'
- − Food photos don't always match the label below them
Stable Diffusion 3.5 Large Turbo
- + Vibrant and artistic food photography
- + High-contrast visual appeal
- − Fails the minimalist layout requirement by creating a messy collage
- − Text is mostly illegible gibberish
- − The 'grid' is inconsistent and floating above separate paper sheets
Verdict: FLUX.2 [pro] successfully created a functional and professional menu design that follows all prompt instructions, including the specific sections and modern aesthetic. Stable Diffusion 3.5 Large Turbo produced high-quality food imagery but failed completely at the structural layout of a menu, resulting in a disorganized collection of elements with unreadable text.
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 [pro]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [pro]
- + Excellent typography rendering for all required text elements.
- + Highly realistic food textures and lighting.
- + Followed all layout instructions including the starburst price tag.
- − Added two patties when the prompt mentioned 'patty' (singular).
Stable Diffusion 3.5 Large Turbo
- + Dynamic sense of motion with fire and embers.
- + Vibrant colors that pop against the dark background.
- − Failed to include any of the requested text.
- − The burger is mostly stacked rather than 'exploded' with suspended components.
- − Lacks the photorealistic detail requested, appearing more illustrative.
Verdict: FLUX.2 [pro] followed the prompt almost perfectly, including the specific text integration and the 'exploded' layout of the burger components. Stable Diffusion 3.5 Large Turbo failed to include any text and produced a mostly intact burger, missing the core structural requirements of the request.
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 [pro]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [pro]
- + Perfect text adherence with 100% accurate spelling of all requested items.
- + Excellent chalk texture and natural handwriting variations.
- + Realistically captured the requested date and pricing details.
- − The 'slate' board has slightly rougher edges than a traditional framed chalkboard, though this fits the 'cozy' aesthetic.
Stable Diffusion 3.5 Large Turbo
- + Clean composition with nice lighting and framing.
- + Includes a warm cafe environment with plants and decor.
- − Numerous spelling errors like 'trulale', 'risctto', and 'ocotpg'.
- − Failed to render the date correctly, showing 'apri 3l 20-'.
- − Text style looks like a digital font rather than authentic chalk handwriting.
Verdict: FLUX.2 [pro] followed the prompt instructions perfectly, rendering every word of the complex menu with flawless spelling and a very convincing handwritten chalk texture. Stable Diffusion 3.5 Large Turbo struggled significantly with the text, producing many gibberish words and failing to capture the specific handwriting style requested.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
FLUX.2 [pro]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [pro]
- + Excellent adherence to the 'horse on top' spatial instruction.
- + Highly detailed cinematic background with multiple galaxies and realistic Earth lighting.
- + Creative interpretation showing a surreal merging of the horse and astronaut.
- − Coherence issue where a second horse body appears to be growing out of the astronaut's back.
Stable Diffusion 3.5 Large Turbo
- + Clear, high-contrast imagery.
- + Good rendering of the horse's anatomy and the astronaut's suit.
- − Completely failed the negative constraint to have the horse on top.
- − Severe lighting inconsistency between the subjects and the environment.
- − The moon/planet in the top left is distorted and low quality.
Verdict: FLUX.2 [pro] followed the difficult spatial instruction to place the horse on top of the astronaut, whereas Stable Diffusion 3.5 Large Turbo defaulted to the common trope of an astronaut riding a horse. Despite some anatomical confusion in the mid-ground of Model A, its composition and adherence to the prompt's surreal and cinematic requirements make it much more successful than Model B.
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 [pro]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [pro]
- + Excellent photorealism with detailed textures on the fur and jacket.
- + Strong prompt adherence including the businesswoman in the back looking at her phone.
- + Highly detailed and realistic taxi interior including dashboard and rain on windows.
- − The capybara's hands look more like human gloves rather than paws.
Stable Diffusion 3.5 Large Turbo
- + Good lighting and bokeh effect in the background.
- + Correct cap and jacket placement on the subject.
- − The businesswoman in the background is blurry and not clearly looking at a phone.
- − The capybara's paws are rendered poorly with anatomically incorrect digits.
- − Overall lighting feels more like a studio render than a photorealistic street scene.
Verdict: FLUX.2 [pro] is the clear winner as it captured every detail of the prompt, including the specific actions of the passenger and a highly complex, realistic interior. Stable Diffusion 3.5 Large Turbo struggled with the paw anatomy and failed to clearly render the businesswoman as requested.
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 [pro]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [pro]
- + Excellent typography with perfect adherence to the required event details.
- + Highly atmospheric and cinematic lighting that fits the 'vintage gothic' theme.
- + Follows all instructions including the scroll banner and specific background elements.
Stable Diffusion 3.5 Large Turbo
- + Clear and bold central jack-o-lantern illustration.
- + Strong graphic border with thorns and webs as requested.
- − Failed to include much of the required text, including the location, date, and time.
- − Missing the scroll banner element.
- − Composition feels more like a generic clip-art frame than a cinematic poster.
Verdict: FLUX.2 [pro] is the clear winner as it perfectly rendered all requested text, including the specific date and location, while maintaining a sophisticated gothic aesthetic. Stable Diffusion 3.5 Large Turbo failed to include several key parts of the prompt, most notably the event details and the scroll banner.
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 [pro]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [pro]
- + Perfect text rendering of 'JAPAN' and 'SUSHI' exactly as requested.
- + Excellent adherence to the 'solid light blue background' and diorama base requirements.
- + Clean, 3D render aesthetic with high clarity and balanced composition.
- − The texture of the rice is a bit stylized/bumpy rather than realistic grains.
Stable Diffusion 3.5 Large Turbo
- + Appealing miniature styling and vibrant colors.
- + Good use of shadows and highlights to create depth in the isometric view.
- − Text is misspelled as 'SIIHI' and the composition is cluttered with wooden signs not in the prompt.
- − The flag icon is incorrect, appearing as a red and white bicolour instead of the Japanese flag.
- − Failed to place text at the top-center as specified.
Verdict: FLUX.2 [pro] followed every instruction perfectly, including the specific positioning and spelling of the text, the background color, and the centered composition. In contrast, Stable Diffusion 3.5 Large Turbo failed on the text rendering, delivered an incorrect flag, and added unnecessary clutter that deviated from the requested minimal aesthetic.
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 [pro]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [pro]
- + Excellent adherence to the prompt, including all four specific animals requested.
- + Hyper-realistic textures on fur and environment with convincing lighting.
- + Highly natural interaction between the animals and the butterflies.
- − The fox's muzzle looks slightly merged with the rabbit's ear area.
Stable Diffusion 3.5 Large Turbo
- + Bright, saturated colors that match a 'wholesome' vibe.
- + Clean, stylized aesthetic following the golden sunrise request.
- − Failed the prompt by missing the baby bunny and the red fox kit.
- − Anatomical issues with the cat's limbs merging into the dog.
- − Visual style is more illustrative/AI-standard than 'hyper-photorealistic'.
Verdict: FLUX.2 [pro] successfully rendered all four requested animals with incredible detail and realism, capturing the soft lighting and intricate fur textures perfectly. Stable Diffusion 3.5 Large Turbo failed to include the bunny and fox, and the overall image quality was less representative of a '8K masterpiece' with several anatomical clipping issues.
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 [pro]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [pro]
- + Perfect adherence to text spelling including the grave accent in 'Caffè'.
- + Clean, professional vector style with high-quality cross-hatching details.
- + Balanced composition with a subtle, realistic paper texture background.
- − The cloche handle is slightly off-center compared to the steam.
Stable Diffusion 3.5 Large Turbo
- + Strong high-contrast lighting creates a bold visual impact.
- + Interesting creative choice to merge the cloche with a coffee mug shape.
- − Includes a spelling error with an extra 'e' in 'Caffeé' and a mangled letter at the end of 'Florian'.
- − Heavy staining effect on the background looks less like 'subtle texture' and more like a messy border.
- − Typography is distorted and lacks the clean alignment expected of a logo.
Verdict: FLUX.2 [pro] followed the prompt more accurately, delivering clean typography and a professional vector aesthetic that matches the 'minimalist logo' requirement perfectly. Stable Diffusion 3.5 Large Turbo struggled with the text rendering, introducing spelling errors and graphical artifacts, and produced a much heavier background texture than 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.2 [pro]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [pro]
- + Excellent typography rendering for main headers and names.
- + Strict adherence to the 6-step infographic structure requested.
- + Perfectly follows the NASA-inspired color palette and flat-vector style.
- − Body text is largely nonsensical 'Lorem Ipsum' style gibberish.
- − The Saturn V icon looks more like a shuttle than a rocket.
Stable Diffusion 3.5 Large Turbo
- + Stronger 'vintage poster' aesthetic with creative layout.
- + High visual interest and good use of space for the main lunar illustration.
- − Poor typography with multiple spelling errors and garbled text.
- − Failed to follow the requested 6-step sequential structure.
- − The iconography is inconsistent and does not match the prompt's specific step list.
Verdict: FLUX.2 [pro] followed the complex layout instructions perfectly, creating a logical 6-step infographic with professional vector icons and accurate names for the crew. Stable Diffusion 3.5 Large Turbo produced a more artistic but disorganized poster that failed to include the requested sequential steps and featured significantly worse text rendering.
Next steps
Explore each model
Distilled version of SD 3.5 Large that generates high-quality images in just 4 steps, offering faster inference and reduced costs