Black Forest Labs' 12 billion parameter distilled image generation model optimized for speed, capable of generating high-quality images in just 4 inference steps
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX.1 [schnell]
18.7
arena score
#48 of 62 in Text-to-Image
Skill signature
· Text-to-Image
Stable Diffusion 3.5 Medium
16.8
arena score
#57 of 62 in Text-to-Image
Vote tally
Where the votes landed
FLUX.1 [schnell]
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.1 [schnell]
Stable Diffusion 3.5 Medium
AI Judge Analysis
FLUX.1 [schnell]
- + Excellent photorealism and texture detail
- + Accurate lighting and reflections consistent with a glass object
- + High resolution with clean, sharp edges
- − Added an extra blue sphere on top of the book not requested in the prompt
- − The internal sphere appears to be floating without a visible support
Stable Diffusion 3.5 Medium
- + Followed the spatial instructions perfectly without adding extra objects
- + Good capture of the 'partially visible through glass' requirement for the plant
- + Nostalgic, soft lighting aesthetic
- − Lower image resolution and visible noise/grain
- − The book texture is somewhat blurry and lacks definition
- − The glass cube edges appear slightly warped
Verdict: Stable Diffusion 3.5 Medium followed the prompt instructions more accurately by placing only one sphere inside the cube, whereas FLUX.1 [schnell] hallucinated a second sphere on top of the book. However, FLUX.1 [schnell] produced a significantly higher quality image with superior lighting, materials, and clarity, making it more visually appealing despite the adherence error.
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 [schnell]
Stable Diffusion 3.5 Medium
AI Judge Analysis
FLUX.1 [schnell]
- + Excellent anatomical detail in the hands and face.
- + The bicycle structure is mechanically coherent and well-rendered.
- + High clarity and sharp focus on the subject with realistic skin texture.
- − Missed the motion blur request for the passing cars.
- − The overall image looks a bit too clean and lacks the requested light rain effect.
- − Pavement looks more damp than wet, with minimal reflections.
Stable Diffusion 3.5 Medium
- + Captures the atmospheric light rain and wet pavement reflections perfectly.
- + Includes cinematic motion blur on the background vehicles as requested.
- + The framing and lighting feel more like a candid, gritty street photograph.
- − The hands and bike mechanics are poorly rendered and distorted.
- − Lower overall resolution and clarity compared to Model A.
- − Bicycle proportions are unrealistic and messy.
Verdict: FLUX.1 [schnell] produces a much higher quality image in terms of detail and subject realism, but it largely ignores the atmospheric requests like motion blur and rain. Stable Diffusion 3.5 Medium follows the stylistic prompts and environmental details much better, but fails significantly on the fine details of the man and the bicycle. FLUX.1 [schnell] is the winner for its superior visual coherence, despite the missed environmental cues.
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 [schnell]
Stable Diffusion 3.5 Medium
AI Judge Analysis
FLUX.1 [schnell]
- + Extremely high skin texture and pore detail
- + Intense and convincing facial expression
- + Bold use of lighting and shadows for a cinematic feel
- − Missed the request for beads in the hair
- − Armor is dark and lacks the requested fine engravings
- − Fails to show the leather straps or cloth underlayer
Stable Diffusion 3.5 Medium
- + Excellent adherence to the 'engraved plate armor' prompt
- + Includes the specific hair braids and bead details
- + Features the requested bokeh sparks and detailed leather straps
- − Skin texture is significantly softer and less lifelike than Image A
- − The facial expression is a bit flat compared to the 'battle-worn' description
- − Lighting feels a bit artificial rather than reflecting naturally off the metal
Verdict: While FLUX.1 [schnell] provides a more visceral and high-fidelity close-up of a face, it ignores several key elements of the prompt like the armor engravings and hair beads. Stable Diffusion 3.5 Medium followed the instruction much more comprehensively, capturing the specific armor style, accessories, and atmospheric effects, even if the skin texture is less detailed.
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 [schnell]
Stable Diffusion 3.5 Medium
AI Judge Analysis
FLUX.1 [schnell]
- + Strong minimalist aesthetic with clean white space.
- + The layout is highly professional and structured.
- + Includes the requested sections for Appetizers, Pizza, and Mains.
- − The 'ORFEFUS' section replaces the requested 'Mains' heading.
- − Contains fewer food photos than expected for a grid layout.
- − Text rendering becomes garbled in the body copy.
Stable Diffusion 3.5 Medium
- + Better variety of food photography in a clear grid pattern.
- + Incorporates more vibrant colors as requested in the prompt.
- + Follows the 'grid' instruction more literally with multiple images.
- − The layout is a bit cluttered, losing the 'minimalist' feel.
- − The font choice is a bit toon-like and lacks the 'bold sans-serif' professional look.
- − Text content is entirely gibberish and symbols.
Verdict: FLUX.1 [schnell] creates a much more realistic and professional layout that feels like an actual restaurant menu, whereas Stable Diffusion 3.5 Medium struggles with typography and aesthetic balance. While Stable Diffusion 3.5 Medium included more images in a grid, FLUX.1 [schnell] follows the minimalist design brief more effectively.
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 [schnell]
Stable Diffusion 3.5 Medium
AI Judge Analysis
FLUX.1 [schnell]
- + Excellent photorealistic texture on the patty, bun, and cheese.
- + Successful incorporation of all text elements including the currency symbol.
- + Dynamic composition with floating crouton-like debris that adds to the 'exploded' feel.
- − Failed the spelling of the primary brand name, writing 'AGIC BURGER' instead of 'MAGIC BURGER'.
- − The text layout is a bit cluttered and overlaps the main subject poorly.
- − The price is repeated twice with confusing formatting.
Stable Diffusion 3.5 Medium
- + Perfectly rendered text for all three requested elements.
- + Very clean and professional advertising layout with a clear starburst.
- + Great sense of heat and atmosphere with the fiery background.
- − Failed the 'exploded' part of the prompt, as the burger is fully assembled rather than having suspended components.
- − The lighting on the burger is a bit flat compared to the intense background fire.
- − Missing the sense of motion and 'suspended in mid-air' individual components.
Verdict: Stable Diffusion 3.5 Medium delivers a much better advertising layout with perfect text rendering, but it completely missed the central prompt requirement for an 'exploded' burger with suspended components. FLUX.1 [schnell] captures the dynamic, exploded concept much better and has superior textures, but it suffers from a typos and a cluttered text design. Stable Diffusion 3.5 Medium is the winner for its professional finish and text accuracy, despite the literal interpretation failure of the physics.
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 [schnell]
Stable Diffusion 3.5 Medium
AI Judge Analysis
FLUX.1 [schnell]
- + Text is highly legible with a clear handwritten feel
- + Correctly captures the casual café atmosphere in the background
- + The layout is clean and follows a logical menu structure
- − Several spelling errors like 'Pril' and 'Octtoopus'
- − The 'elegant cursive' requirement for the title was not met, using block letters instead
- − Included repetitive and nonsensical text at the bottom
Stable Diffusion 3.5 Medium
- + Excellent chalk texture with realistic smudging and dusting
- + Creative interpretation of the chalkboard layout
- + Attempts a more decorative lettering style similar to what was requested
- − Text is largely illegible with severe garbling of words
- − Failed significantly on the date and specific pricing requests
- − The composition feels cluttered and chaotic
Verdict: FLUX.1 [schnell] is the clear winner because it produces legible, semi-accurate text that follows the intent of a menu, despite some spelling errors and a lack of cursive. Stable Diffusion 3.5 Medium creates a very realistic chalk texture, but the text is completely broken and fails to convey the specific information requested in the prompt.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
FLUX.1 [schnell]
Stable Diffusion 3.5 Medium
AI Judge Analysis
FLUX.1 [schnell]
- + Perfectly follows the specific instruction of having the horse on top of the astronaut
- + Strong cinematic lighting with a glowing planetary background
- + High level of texture detail on the astronaut suit and horse fur
- − The horse appears to have two heads emerging from the same neck
- − Anatomically confusing connection between the horse and the astronaut's equipment
Stable Diffusion 3.5 Medium
- + Crisp and clear image quality
- + Good composition with the horse walking across the horizon
- + Accurate astronaut suit rendering
- − Completely failed the negative constraint to have the horse on top
- − The horse has severe anatomical glitches including six legs
Verdict: FLUX.1 [schnell] is the clear winner as it successfully interpreted the challenging spatial prompt to place the horse on top of the astronaut. Stable Diffusion 3.5 Medium defaulted to the common concept of an astronaut riding a horse and suffered from significant anatomical errors, generating a six-legged horse.
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 [schnell]
Stable Diffusion 3.5 Medium
AI Judge Analysis
FLUX.1 [schnell]
- + Excellent character placement and anatomy, with the capybara's paw naturally resting on the steering wheel.
- + The passenger is clearly interacting with a phone as requested, with a realistic bored expression.
- + High-quality fur texture and clean text rendering on the taxi hat.
- − The passenger appears to be in the front seat next to the driver rather than the back seat.
Stable Diffusion 3.5 Medium
- + Correctly positions the passenger in the back seat as specified in the prompt.
- + Detailed taxi driver cap with a realistic-looking badge.
- + Good bokeh effect for the Manhattan street lights through the window.
- − The passenger is not looking at her phone; her hands are empty and she is looking ahead.
- − The capybara's anatomy is slightly confused around the steering wheel area, with paws appearing cluttered.
- − The lighting on the capybara's face is somewhat flat compared to the passenger.
Verdict: FLUX.1 [schnell] captures the specific action of the passenger looking at her phone much better than Stable Diffusion 3.5 Medium, although it places her in the front seat. FLUX.1 [schnell] also produces a much more convincing integration of animal anatomy with the car's controls, whereas Stable Diffusion 3.5 Medium succeeds at the spatial layout but fails on the key character action.
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 [schnell]
Stable Diffusion 3.5 Medium
AI Judge Analysis
FLUX.1 [schnell]
- + Features a balanced central composition with cinematic lighting
- + Good interpretation of gothic thorns and web details in the border
- + Overall artistic style is polished and visually appealing
- − Included redundant and garbled text lines at the bottom
- − Grammar error in the subhead ('You invited to a a night')
- − The 'parchment' texture is less distinct compared to Model B
Stable Diffusion 3.5 Medium
- + Strong parchment effect for the central poster element
- + Vibrant colors on the jack-o-lanterns
- + Better execution of the twisted trees in the background
- − Major spelling errors in several prominent words including 'Halloweeen' and 'Inviloween'
- − Repeated typos in address and date details
- − Layout feels cramped with inconsistent text alignment
Verdict: FLUX.1 [schnell] is the winner because it adheres much closer to the requested text and maintains a cleaner, more professional composition. Stable Diffusion 3.5 Medium struggled significantly with spelling, producing gibberish for the main title and failing to follow the specific date and location strings accurately.
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 [schnell]
Stable Diffusion 3.5 Medium
AI Judge Analysis
FLUX.1 [schnell]
- + Perfectly captures the miniature diorama aesthetics with a distinct raised base.
- + Clean, professional graphic design and typography for the 'JAPAN' text and flag.
- + Excellent simulation of PBR materials with soft, realistic lighting.
- − Failed to include the sub-text 'SUSHI' requested in the prompt.
- − The sushi roll cross-section looks slightly anatomically confused compared to real sushi.
Stable Diffusion 3.5 Medium
- + Included both 'JAPAN' and 'SUSHI' text as requested.
- + Good texture on the nori and roe, providing a nice contrast.
- + Effective use of the light blue background.
- − Failed to include the requested flag icon.
- − The 'diorama base' is just a standard plate rather than a raised miniature stage.
- − The typography is cluttered and overlapping rather than being 'large bold' and 'below it'.
Verdict: FLUX.1 [schnell] produced a far superior visual result that perfectly nailed the 'miniature 3D cartoon scene' and 'raised diorama' aesthetic, though it missed the second line of text. Stable Diffusion 3.5 Medium followed the text instruction better but failed on the 3D diorama composition and the flag icon, resulting in a flatter, less professional-looking image.
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 [schnell]
Stable Diffusion 3.5 Medium
AI Judge Analysis
FLUX.1 [schnell]
- + Excellent adherence to the count and variety of animals requested.
- + Effective use of warm golden lighting and 'god rays'.
- + Characters appear more integrated into the floral environment.
- − One animal appears as a hybrid between a kitten and a bunny.
- − Image has a slightly soft, digital-painting feel rather than true hyper-photorealism.
Stable Diffusion 3.5 Medium
- + High textural detail in the fur of the golden retriever and fox.
- + Vibrant color palette with crisp focus on the foreground subjects.
- − Failed to include the requested baby bunny, showing only three animals.
- − The lighting feels a bit artificial and overly saturated.
- − Anatomical issues where the kitten's front paws blend unnaturally into the grass.
Verdict: FLUX.1 [schnell] followed the complex prompt more accurately by including all four requested animal types, whereas Stable Diffusion 3.5 Medium omitted the baby bunny entirely. While Stable Diffusion 3.5 Medium offered sharper fur textures, FLUX.1 [schnell] captured the 'wholesome' sunrise atmosphere much more effectively, despite the biological ambiguity of the kitten-bunny hybrid.
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 [schnell]
Stable Diffusion 3.5 Medium
AI Judge Analysis
FLUX.1 [schnell]
- + Clean vector aesthetic suitable for a logo
- + Minimalist layout that matches the prompt style
- + High resolution with sharp typography
- − Significant spelling errors in every piece of text
- − Date is incorrectly rendered as 7720
- − Missing the steam element requested
Stable Diffusion 3.5 Medium
- + More intricate vintage illustration style
- + Closer to the requested name spelling
- + Captures an artistic, textured felt-tip aesthetic
- − Fails to follow the 'minimalist' instruction
- − Typography is messy and contains spelling errors
- − Incorrect establishment date and illegible text on the banner
Verdict: Both models struggled significantly with the text requirements, which is the core of logo design. FLUX.1 [schnell] produced a much cleaner vector layout that fits the 'minimalist' prompt perfectly, but failed entirely on the spelling of 'Florian' and the date. Stable Diffusion 3.5 Medium provided a more detailed vintage illustration, but the result is overly cluttered and equally flawed in its text rendering. FLUX.1 [schnell] is preferred as a starting point for a logo, despite the typos, due to its superior composition and professional finish.
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 [schnell]
Stable Diffusion 3.5 Medium
AI Judge Analysis
FLUX.1 [schnell]
- + Excellent layout that uses the entire canvas to tell a narrative story.
- + Clean, professional vector aesthetic that perfectly matches the requested flat style.
- + High symmetry and visual balance in the central composition.
- − The main rocket icon looks a bit like a cartoon shuttle rather than a Saturn V.
- − Text is mostly illegible gibberish, failing to provide specific step info.
Stable Diffusion 3.5 Medium
- + Includes 6 distinct steps corresponding to the prompt requirements.
- + Good color palette adherence with the navy, white, and red combo.
- + Clearer text rendering for some keywords like 'Launch' and 'Landing'.
- − The composition is cluttered and cut off on the right side.
- − Iconography is inconsistent, with some steps being wireframe globes and others silhouettes.
- − Does not follow the flat-vector style well, appearing more like a textured digital painting.
Verdict: FLUX.1 [schnell] creates a much more cohesive and visually appealing vector poster with professional symmetry, although the text is mostly unreadable. Stable Diffusion 3.5 Medium attempts to follow the 6-step structure more literally, but the resulting composition is awkward and the art style is inconsistent. FLUX.1 [schnell] is the winner for its superior design sense and adherence to the requested flat-vector aesthetic.
Next steps
Explore each model
Stability AI's 2.5-billion parameter Multimodal Diffusion Transformer with improvements (MMDiT-X) text-to-image model optimized for consumer hardware, featuring improved image quality, typography, and complex prompt understanding