Black Forest Labs' enhanced 12-billion parameter flow transformer with 6x faster generation than FLUX.1 [pro], delivering superior composition, detail, and artistic fidelity
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX1.1 [pro]
18.4
arena score
#50 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
FLUX1.1 [pro]
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.”
FLUX1.1 [pro]
Stable Diffusion 3.5 Large
AI Judge Analysis
FLUX1.1 [pro]
- + Perfect adherence to the spatial requirements of the prompt.
- + Highly realistic textures on the book and wooden table.
- + Sophisticated lighting with accurate reflections on the glass and sphere surface.
- − The glass enclosure is more of a tall rectangular prism than a perfect cube.
Stable Diffusion 3.5 Large
- + Realistic handling of dust and smudges on the glass surface.
- + Good lighting and shadows that match the environment.
- − Failed the spatial prompt: the red book is inside the cube rather than on top of it.
- − The blue sphere is sitting on the book instead of just being 'inside the cube' independently.
- − The green plant is predominantly above the cube rather than 'behind it' as requested.
Verdict: FLUX1.1 [pro] followed every instruction perfectly, placing the red book on top and the sphere inside while maintaining a high level of photorealism. Stable Diffusion 3.5 Large failed the core spatial reasoning of the prompt, placing the book underneath the sphere inside the cube.
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.”
FLUX1.1 [pro]
Stable Diffusion 3.5 Large
AI Judge Analysis
FLUX1.1 [pro]
- + Excellent handling of wet pavement reflections and bokeh
- + Character skin texture and clothing look very realistic
- + Captured the 'candid' and 'imperfect framing' request effectively
- − Severe anatomical errors with the bicycle including missing rear parts
- − The man's feet and the objects on the ground are poorly defined and fused
Stable Diffusion 3.5 Large
- + Successfully rendered the full bicycle structure
- + Effective use of rain-streak effects to convey weather
- + Strong adherence to the 'Japanese man' and 'red bicycle' descriptors
- − Lacks the requested 'motion blur from passing cars'
- − The skin texture and hair look slightly more artificial/painterly compared to A
- − Cars in the background are too sharp, missing the shallow depth of field effect
Verdict: While FLUX1.1 [pro] excels at the photographic texture, lighting, and cinematic atmosphere requested, it fails significantly on the structural integrity of the bicycle and the man's lower body. Stable Diffusion 3.5 Large provides a more coherent and complete scene with a realistic bicycle, though it missed specific technical prompts like motion blur and shallow depth of field. Stable Diffusion 3.5 Large is the preferred choice here because the structural failures in FLUX1.1 [pro] are too distracting for a realistic photo.
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.”
FLUX1.1 [pro]
Stable Diffusion 3.5 Large
AI Judge Analysis
FLUX1.1 [pro]
- + Exceptional skin texture with realistic pores and grime
- + Dynamic use of bokeh sparks and warm lighting
- + Intense, lifelike gaze with high clarity in the eyes
- − Hair braids are messy and poorly defined
- − Missing the beads mentioned in the prompt
- − Leather and cloth textures are mostly obscured by the tight framing
Stable Diffusion 3.5 Large
- + Excellent adherence to the 'braided hair' and 'ornate engraved armor' requirements
- + Great visibility of the cloth underlayer and mail
- + Strong cinematic composition showing a larger portion of the character
- − Lighting is a bit flat compared to the 'warm torchlight' request
- − Skin and facial hair look slightly more synthetic than the competitor
- − Misses the small beads in the braids
Verdict: FLUX1.1 [pro] produces a much more realistic and intimate portrait with superior skin textures and lighting effects. Stable Diffusion 3.5 Large follows the armor and hairstyle instructions more closely but lacks the lifelike fidelity and atmospheric lighting found in the FLUX output.
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.”
FLUX1.1 [pro]
Stable Diffusion 3.5 Large
AI Judge Analysis
FLUX1.1 [pro]
- + Excellent presentation of the menu as a physical object with realistic utensils.
- + Features a well-structured grid for food photography within the layout.
- + Includes distinct sections for 'Appetizers', 'Pizza', and 'Mains' as requested.
- − The text varies in clarity, with some illegible symbols.
- − The photography grid is localized to the bottom rather than integrated through the layout.
Stable Diffusion 3.5 Large
- + Striking and bold sans-serif typography that feels modern.
- + High-quality, vibrant food photography integrated as a border grid.
- + Very clean minimalist aesthetic that adheres to the white background requirement.
- − The text has significant spelling errors in the headers (e.g., 'MAIMAES', 'APPETIZRS').
- − Does not show the full menu context/mockup, focusing only on the digital-style layout.
Verdict: FLUX1.1 [pro] provides a more complete 'menu design' package by creating a realistic mockup with cutlery and clear, distinct sections, although its text rendering is less bold. Stable Diffusion 3.5 Large offers a cleaner, more vibrant aesthetic and follows the grid requirement more literally, but it suffers from significant spelling errors and hallucinatory characters. FLUX1.1 [pro] is the better choice for a professional layout that balances photography and text functionality.
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.”
FLUX1.1 [pro]
Stable Diffusion 3.5 Large
AI Judge Analysis
FLUX1.1 [pro]
- + Excellent typography rendering for all requested text elements.
- + Clean composition that balances the product with the secondary fragments.
- + Strong adherence to the price and currency request.
- − The burger is not truly 'exploded' into individual components as requested; it is mostly intact.
- − The price is missing the starburst element.
- − The repetition of the brand name in two different styles is slightly redundant.
Stable Diffusion 3.5 Large
- + Atmospheric fire effects that create a high-impact visual.
- + High-quality texture on the burger patties and melting cheese.
- + Dynamic lighting that matches the fiery background.
- − Failed to include any of the requested text elements ('MAGIC BURGER', price, etc.).
- − Ignored the 'exploded burger' requirement, showing a fully assembled burger.
- − Missing the starburst and specific fiery/glowing text effects.
Verdict: FLUX1.1 [pro] is the clear winner as it successfully integrated almost all complex text requirements and pricing, which Stable Diffusion 3.5 Large completely failed to render. While FLUX1.1 [pro] leaned toward a floating burger rather than a fully exploded one, it followed the 'ad' format instructions much better than the text-less output of Stable Diffusion 3.5 Large.
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.”
FLUX1.1 [pro]
Stable Diffusion 3.5 Large
AI Judge Analysis
FLUX1.1 [pro]
- + Excellent text legibility and alignment.
- + Accurate date rendering with '2026' as requested.
- + Clean, professional composition with realistic chalk texture.
- − Hallucinated prices and repeated/split items (Grilled Octopus listed twice with wrong prices).
- − Text rendering is slightly too clean, appearing somewhat digital rather than organic chalk handwriting.
Stable Diffusion 3.5 Large
- + Natural, messy chalk aesthetic that feels authentically handwritten.
- + Great environmental context with the cozy café interior.
- − Numerous spelling errors including 'Todaay', 'Ottpups', and 'Cholcalte'.
- − Failed to follow the specific date '2026', using '2024' instead.
- − Layout is cluttered and difficult to read.
Verdict: FLUX1.1 [pro] followed the complex text instructions much better, correctly rendering the specific date and most item names, although it struggled with the logical consistency of prices and quantity of items. Stable Diffusion 3.5 Large captured a more authentic 'chalk' feel and café atmosphere, but failed significantly on spelling and specific prompt details like the year and item list.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
FLUX1.1 [pro]
Stable Diffusion 3.5 Large
AI Judge Analysis
FLUX1.1 [pro]
- + Strong cinematic lighting with dramatic contrast
- + High level of detail in the horse's anatomy and astronaut suit
- − The environment resembles a planetary surface more than open space
- − The horse's mechanical-looking torso creates a confusing hybrid design
Stable Diffusion 3.5 Large
- + Excellent adherence to the 'in space' setting with a galactic background
- + Graceful composition that conveys movement through nebulae
- − Lower resolution and clarity in the fine details compared to Model A
- − The horse's anatomy appears somewhat distorted around the hind legs
Verdict: Both models successfully interpreted the prompt, but FLUX1.1 [pro] produced a more polished, high-definition image with superior texturing. Stable Diffusion 3.5 Large captured the 'space' atmosphere better, though it suffered from softer details and anatomical inconsistencies.
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.”
FLUX1.1 [pro]
Stable Diffusion 3.5 Large
AI Judge Analysis
FLUX1.1 [pro]
- + Excellent atmospheric lighting and bokeh
- + Perfectly captures the passenger looking at her phone as requested
- + Strong adherence to the 'bored expression' prompt
- − The steering wheel is mostly obscured, missing the 'both paws on wheel' detail
- − The human passenger appears more glamorous than a 'bored businesswoman'
Stable Diffusion 3.5 Large
- + Successfully places paws on the steering wheel
- + Very detailed high-contrast textures on the fur and leather jacket
- + Strong yellow taxi color saturation and clear NYC street lighting
- − Completely failed to include the human businesswoman in the back seat
- − The capybara's anatomy is slightly distorted with a more rodent-like/mouse-like ear shape
- − The view is from outside looking in, rather than 'inside the taxi'
Verdict: FLUX1.1 [pro] followed the complex narrative elements of the prompt much more successfully by including both the capybara driver and the bored passenger, creating a cohesive scene. Stable Diffusion 3.5 Large produced a high-quality portrait of a capybara but entirely missed the second character and the 'inside' perspective requested, resulting in a composition that failed half of the prompt's instructions.
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.”
FLUX1.1 [pro]
Stable Diffusion 3.5 Large
AI Judge Analysis
FLUX1.1 [pro]
- + Excellent visual coherence and moody lighting
- + Accurate inclusion of specific event details like the date and location
- + Highly detailed thorny border that feels polished
- − Repeat text artifacts where lines of information are duplicated
- − The banner text contains a spelling error ('univted')
Stable Diffusion 3.5 Large
- + Strong composition with a prominent, classic scroll element
- + Text on the scroll is clear and correctly spelled
- + Better adherence to the 'parchment' texture request
- − Completely failed to include the event details at the bottom (Date, Time, Location)
- − The 'Central jack-o-lantern' is off to the side, not central
- − The moon is overly blown out, losing some gothic subtlety
Verdict: FLUX1.1 [pro] produced a more complete invitation with all requested text data, despite some duplication and minor spelling errors. In contrast, Stable Diffusion 3.5 Large failed to include the event details entirely and missed the 'central' placement of the jack-o-lantern, even though its parchment texture and scroll design were visually appealing.
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.”
FLUX1.1 [pro]
Stable Diffusion 3.5 Large
AI Judge Analysis
FLUX1.1 [pro]
- + Excellent adherence to the '3D cartoon scene' and 'miniature' style instructions.
- + Clean, professional typography that is perfectly centered as requested.
- + Smooth, refined textures that match the soft PBR aesthetic.
- − Omitted the requested flag icon.
- − The background is very sparse, bordering on plain.
Stable Diffusion 3.5 Large
- + Successfully included all elements, including the small flag icon.
- + High level of detail on the sushi ingredients and rice texture.
- + Includes creative additions like chopsticks and a side dish.
- − Failed the centering instruction for text, placing it on a sign instead of top-center.
- − Texture is more realistic/gritty than the requested 'soft refined' cartoon style.
- − Composition feels cluttered compared to the 'minimal garnish' request.
Verdict: FLUX1.1 [pro] better captured the requested isometric cartoon aesthetic and clean typography placement, resulting in a more cohesive diorama. Stable Diffusion 3.5 Large followed more individual prompts like the flag icon but failed on the layout/centering constraints and the overall 'soft cartoon' stylistic tone.
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.”
FLUX1.1 [pro]
Stable Diffusion 3.5 Large
AI Judge Analysis
FLUX1.1 [pro]
- + Exceptional backlighting and rim lighting on the fur
- + High clarity and sharp focus on the animals' faces
- + Artistically consistent butterfly elements
- − Missed one of the four requested animals (no red fox kit)
- − Animals are sitting rather than 'playfully chasing' or 'tumbling'
Stable Diffusion 3.5 Large
- + Included all four requested animals: puppy, kitten, bunny, and fox kit
- + Captures the action of 'playfully chasing' with dynamic poses
- + Beautiful rendering of dew sparkles and god rays
- − The fox kit's face looks slightly more artificial/cartoonish than the others
- − Some minor blending artifacts where the fox and puppy overlap
Verdict: Stable Diffusion 3.5 Large is the winner because it successfully included all four requested animals and captured the 'chasing' action described in the prompt. While FLUX1.1 [pro] offered slightly more refined lighting and fur textures, it failed the prompt adherence test by omitting the fox kit and depicting the animals in a static pose.
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.”
FLUX1.1 [pro]
Stable Diffusion 3.5 Large
AI Judge Analysis
FLUX1.1 [pro]
- + Excellent classical engraving aesthetic that fits the 'Vintage' prompt perfectly.
- + Beautiful border details and consistent brown/cream color palette.
- + High visual quality with sophisticated linework and texture.
- − Failed the spelling of the requested brand name, rendering 'Cafeé FRATILIAN' instead of 'Caffè Florian'.
- − The banner does not contain the 'Est. 1720' text as requested; that text is placed below it instead.
Stable Diffusion 3.5 Large
- + Successfully spelled the name 'Caffé Florian' (minor accent variation).
- + Accurately placed 'Est. 1720' in a banner-adjacent design.
- + Clean vector-style icons that match the minimalism requested.
- − The 'cloche' illustration is awkward and disjointed with overlapping steam elements.
- − The composition feels slightly bottom-heavy and less unified than Model A.
Verdict: FLUX1.1 [pro] produced a much more professional and aesthetically pleasing vintage logo but failed significantly on the text rendering, misspelling 'Florian' as 'Fratilian'. Stable Diffusion 3.5 Large adhered much better to the text requirements and banner placement, though its cloche illustration is visually messy compared to the elegant engraving of its competitor.
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.”
FLUX1.1 [pro]
Stable Diffusion 3.5 Large
AI Judge Analysis
FLUX1.1 [pro]
- + Excellent adherence to the clean, flat-vector infographic style.
- + Clear, readable central title and numbered sequence.
- + Effective use of the requested NASA-inspired color palette.
- − Step numbering is repetitive and confusing (multiple '1's, missing '5').
- − Text descriptions for steps are mostly illegible gibberish.
Stable Diffusion 3.5 Large
- + Includes more scientific-looking textural detail on planet surfaces.
- + Captures the 'modern infographic' vibe with a complex horizontal layout.
- − Failed to provide a clear numbered 6-step sequence as requested.
- − The 'Saturn V' icon looks more like a modern Space Shuttle with boosters.
- − The composition is cluttered with significant amounts of illegible placeholder text.
Verdict: FLUX1.1 [pro] better adheres to the specific request for a numbered 6-step diagram in a flat-vector style, even though it struggles with the logic of the numbering. Stable Diffusion 3.5 Large creates a more visually complex poster, but it fails to follow the structural instructions of the prompt and depicts the wrong spacecraft for the Apollo missions.
Next steps
Explore each model
Stability AI's 8.1-billion parameter Multimodal Diffusion Transformer (MMDiT) text-to-image model featuring improved image quality, typography, complex prompt understanding, and resource-efficiency