Black Forest Labs' precision image generation model with maximum control, reliable text rendering, and complete creative control supporting up to 4MP output
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX.2 [flex]
24.8
arena score
#14 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 [flex]
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 [flex]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [flex]
- + Perfect adherence to spatial instructions with the book correctly on top and sphere inside.
- + Highly realistic rendering of glass refraction and soft window lighting.
- + Logical composition with the plant correctly positioned behind the objects.
- − The sphere is quite large relative to the cube, making the 'small blue sphere' prompt element slightly less accurate.
Stable Diffusion 3.5 Large Turbo
- + Sharp details on the wooden texture and leaves.
- + Dynamic lighting with clear shadows cast from left to right.
- − Failed spatial reasoning as the book is inside the cube rather than on top.
- − The plant is to the side rather than behind the cube.
- − The glass cube features a black frame which was not requested.
Verdict: FLUX.2 [flex] followed every single spatial instruction perfectly, correctly placing the book on top of the cube and the sphere inside. Stable Diffusion 3.5 Large Turbo struggled with the layout, placing the book inside the cube and the plant to the left instead of behind.
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 [flex]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [flex]
- + Excellent adherence to the 'imperfect framing' and 'candid' street photography style.
- + Highly realistic skin textures and fine details on the hands and hair.
- + Effective use of motion blur and bokeh in the background to create a cinematic feel.
- − The structural anatomy of the bicycle frame is slightly nonsensical near the pedals.
Stable Diffusion 3.5 Large Turbo
- + Accurately depicts the red bicycle and the light rain effect.
- − The subject's head and neck anatomy are strangely distorted and lack realism.
- − The image appears flat and lacks the requested depth of field and motion blur.
- − The textures are plastic-like and lack the requested natural skin detail.
Verdict: FLUX.2 [flex] successfully captures the requested cinematic photography style with realistic lighting, textures, and atmospheric effects. Stable Diffusion 3.5 Large Turbo fails on nearly every aesthetic prompt requirement, producing a flat image with significant anatomical distortions in the subject's posture.
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 [flex]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [flex]
- + Excellent adherence to all prompt details including ornate engraving and beads in hair.
- + High realism with lifelike skin texture and convincing depth of field.
- + Dynamic lighting that realistically reflects off the metal and lens.
- − The scars appear more like fresh cuts rather than 'faint scars' as specified.
Stable Diffusion 3.5 Large Turbo
- + Strong composition with a focused, intense gaze.
- + Included braided hair and ornate armor elements.
- − Skin texture is overly smooth and looks like plastic or a digital painting.
- − Missed several prompt details such as the beads in the hair and detailed leather/cloth textures.
- − Lighting is harsh and contains digital artifacts around the hair and edges.
Verdict: FLUX.2 [flex] produced a much more realistic and detailed image that followed the prompt's technical requirements for texture and jewelry. Stable Diffusion 3.5 Large Turbo lacks the fine detail and realism requested, resulting in a somewhat artificial, plastic-like appearance for the skin and hair.
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 [flex]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [flex]
- + Excellent typography rendering with readable fonts and prices.
- + Follows the grid layout and section requirements perfectly.
- + Professional graphic design aesthetic suitable for a real menu.
- − The placeholder text for descriptions is gibberish.
- − The 'Appetizers' header is repeated twice in small text below the main heading.
Stable Diffusion 3.5 Large Turbo
- + Creative use of top-down food photography.
- + High-contrast, vibrant colors that make the food stand out.
- − Fails the menu design format, looking more like a scattered flat-lay than a graphic layout.
- − Text is completely illegible and contains multiple spelling errors like 'Mians'.
- − The grid photography doesn't integrate well with the text sections.
Verdict: FLUX.2 [flex] successfully interprets the prompt as a graphic design task, producing a clean, professional-looking menu with a clear grid and logical sections. In contrast, Stable Diffusion 3.5 Large Turbo fails to provide an integrated layout, offering scattered elements with illegible text and poor spelling.
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 [flex]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [flex]
- + Excellent photorealistic texture and detail on the food components.
- + Accurate rendering of all requested text elements with high legibility.
- + Great sense of motion with crumbs and sauce droplets suspended naturally.
- − The 'starburst' for the price is a bit generic and clip-art like compared to the burger realism.
Stable Diffusion 3.5 Large Turbo
- + Dynamic use of fire and smoke for a dramatic atmosphere.
- + Strong color contrast and vibrant lighting.
- − Failed to include any of the requested text: 'MAGIC BURGER', 'LIMITED TIME ONLY', or price.
- − The burger is not truly 'exploded' with separated layers; it remains mostly stacked.
- − The bottom bun looks distorted and melting rather than photorealistic.
Verdict: FLUX.2 [flex] successfully followed every instruction in the prompt, including complex text rendering and precise arrangement of the 'exploded' layers. Stable Diffusion 3.5 Large Turbo failed to include any text and produced a less realistic image that felt more illustrative than photorealistic.
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 [flex]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [flex]
- + Excellent text rendering with perfect spelling and realistic chalk texture.
- + Authentic chalkboard aesthetic with smudges and natural variations in stroke.
- + Followed all specific menu items and date details perfectly.
- − The slant in handwriting is very uniform, leaning slightly toward a digital font look in some letters.
Stable Diffusion 3.5 Large Turbo
- + Bright, modern cafe setting with nice plant accents.
- + Good lighting and shadows.
- − Severe spelling errors throughout the text (e.g., 'apri 3l 20-', 'trulale', 'ocotpgs').
- − Failed to follow the specific menu items requested.
- − The text looks like a clean vector brush rather than realistic chalk.
Verdict: FLUX.2 [flex] produced a near-perfect result, successfully rendering every complex word and date exactly as requested with a convincing chalk-on-board texture. In contrast, Stable Diffusion 3.5 Large Turbo struggled with basic spelling and failed to follow the specific prompt instructions for the menu items, resulting in a low-quality output for this text-heavy task.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
FLUX.2 [flex]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [flex]
- + Excellent adherence to the 'horse on top' spatial instruction
- + Highly detailed and cinematic background with nebulae and asteroids
- + Superior rendering of textures on the horse's coat and space suit
- − The horse's front legs are awkwardly merged with the astronaut's shoulders
- − Anatomy of the horse's belly is slightly distorted where it meets the astronaut
Stable Diffusion 3.5 Large Turbo
- + Clear, clean lighting and sharp contrast
- + Good rendering of the astronaut suit and saddle details
- − Failed the negative constraint; the astronaut is riding the horse instead of vice versa
- − The horse's hooves are poorly formed and blurry
- − Composition is generic and less surreal than requested
Verdict: FLUX.2 [flex] successfully followed the difficult spatial prompt, placing the horse on top of the astronaut in a creative, surreal manner. Stable Diffusion 3.5 Large Turbo completely failed the core instruction of the prompt, providing a standard 'astronaut on horse' image with significant blurring in the horse's lower limbs.
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 [flex]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [flex]
- + Excellent adherence to the 'human in the back seat' prompt with a clear bored expression and phone usage.
- + Highly realistic taxi interior and exterior background textures.
- + Anatomically convincing capybara with realistic fur texture and human-like paws on the wheel.
- − The capybara's hands are a bit too human-like, which may be slightly unsettling.
- − The side window frame seems to vanish into the car interior.
Stable Diffusion 3.5 Large Turbo
- + Good lighting on the capybara's face and whiskers.
- + Dynamic bokeh effect in the background lights.
- − Failed to include the human businesswoman looking at her phone in the back seat as requested.
- − The capybara's texture is somewhat stylized and less photorealistic than Image A.
- − The steering wheel appears to be floating or incorrectly attached to the dashboard.
Verdict: FLUX.2 [flex] is the clear winner as it followed every part of the prompt, including the specific character in the back seat and her action. Stable Diffusion 3.5 Large Turbo failed to include the businesswoman and had several structural issues with the steering wheel and car 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 [flex]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [flex]
- + Excellent typography rendering for all requested text fields.
- + Superb atmospheric lighting and realistic textures.
- + Perfect adherence to all prompt elements including the scroll banner and specific event details.
- − The parchment edges are slightly repetitive in their tattered pattern.
Stable Diffusion 3.5 Large Turbo
- + Strong composition with a clear central focus.
- + Graphic, clean illustrated style.
- − Failed to include the majority of the requested text, including the scroll banner and event details.
- − The 'Party' text contains a typo ('Partyj').
- − Lighting feels more like a 2D illustration rather than 'cinematic' as requested.
Verdict: FLUX.2 [flex] achieved a near-perfect result, successfully rendering all the complex text requirements and maintaining a highly polished, atmospheric gothic aesthetic. Stable Diffusion 3.5 Large Turbo failed significantly in prompt adherence, missing the specific event details and including a spelling error.
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 [flex]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [flex]
- + Excellent typography rendering with perfect spelling.
- + Accurate representation of the Japanese flag icon.
- + Strict adherence to the light blue background and centralized composition.
- − The diorama base is a bit plain compared to the sushi detail.
Stable Diffusion 3.5 Large Turbo
- + Strong 3D miniature aesthetic with nice material textures on the rice.
- + Includes a physical flag and sign within the diorama scene.
- − Typos in the text, rendering 'SUSHI' as 'SIIHI'.
- − The flag icon is incorrect (red and white stripes/blocks rather than the Japanese flag).
- − Failed to place the text centered at the top as requested.
Verdict: FLUX.2 [flex] adhered much more closely to the specific layout instructions, correctly spelling the text and following the flag icon request. Stable Diffusion 3.5 Large Turbo had better textured 'clay-like' sushi, but failed significantly on text accuracy and the specific icon request.
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 [flex]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [flex]
- + Strictly follows the prompt by including all four specified animals: puppy, kitten, bunny, and fox kit.
- + Excellent environmental details, including realistic god rays, dew on the grass, and a variety of wildflowers.
- + Highly realistic fur textures and anatomy that align with the 'hyper-photorealistic' request.
- − The fox kit's hind leg positioning looks slightly awkward during its running pose.
Stable Diffusion 3.5 Large Turbo
- + Vibrant, eye-catching colors and clean backlighting on the animals' fur.
- + Expressive and large eyes that give the characters an adorable, illustrative quality.
- − Failed to include the bunny and fox kit, only rendering two kittens and a puppy.
- − The puppy's ears and head shape are anatomically inconsistent, and the overall style is more '3D render' than 'photorealistic'.
- − Extra paws are visible in the center, protruding from the animals incorrectly.
Verdict: FLUX.2 [flex] is the clear winner as it successfully rendered all four requested animals with high anatomical accuracy and beautiful atmospheric lighting. Stable Diffusion 3.5 Large Turbo failed the prompt by missing two of the four animals and introduced significant anatomical artifacts like extra limbs.
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 [flex]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [flex]
- + Excellent typography with correct Italian diacritics
- + Perfect adherence to the minimalist vector emblem style
- + Clean and balanced composition
- − Subtle texture is almost invisible, appearing mostly as a solid color
Stable Diffusion 3.5 Large Turbo
- + Strong vintage character with great use of texture and shading
- + Excellent rendering of the 'Est. 1720' banner
- − Misspelled the primary brand name as 'Caffeé Florin'
- − Included a cup handle that was not in the prompt, complicating the cloche silhouette
- − The typography is slightly messy and inconsistent in the main brand name
Verdict: FLUX.2 [flex] produced a professional, clean logo that perfectly followed the typography and style requirements, whereas Stable Diffusion 3.5 Large Turbo struggled with spelling (adding an extra 'e' and dropping the 'a' in Florian) and included extraneous visual elements. FLUX.2 [flex] is the clear winner for its functional design and accuracy.
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 [flex]
Stable Diffusion 3.5 Large Turbo
AI Judge Analysis
FLUX.2 [flex]
- + Excellent typography with clean, legible text including specific mission stages.
- + Perfect adherence to the requested flat-vector style and NASA-inspired color palette.
- + Logical infographic flow that clearly represents the chronological steps of the mission.
- − Misses the final 'Landing' step on the lunar surface, stopping at 'Lunar Orbit' and 'Descent'.
- − The layout feels a bit sparse with significant empty space in the center.
Stable Diffusion 3.5 Large Turbo
- + Strong aesthetic composition with a classic retro-future poster feel.
- + Good use of the color palette to create a cohesive visual theme.
- + Includes a clear depiction of the lunar surface.
- − Poor text rendering with numerous misspellings like 'Apoll.o' and 'Laurch'.
- − Fails to follow the requested 6-step infographic structure, offering nonsensical icons and labels.
- − The iconography is inconsistent and does not clearly represent the mission stages.
Verdict: FLUX.2 [flex] is the clear winner as it successfully creates a functional, legible infographic that follows almost all technical and stylistic requirements. While it missed the very last step of the sequence, it provided crisp, accurate vector icons and readable text. Stable Diffusion 3.5 Large Turbo failed at both legibility and structured adherence, producing garbled text and a generic poster layout rather than the requested step-by-step infographic.
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