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 Medium
16.8
arena score
#57 of 62 in Text-to-Image
Vote tally
Where the votes landed
FLUX.2 [flex]
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 [flex]
Stable Diffusion 3.5 Medium
AI Judge Analysis
FLUX.2 [flex]
- + Excellent photographic realism and lighting
- + Matches all spatial prompt requirements perfectly
- + Realistic glass refraction and depth of field
- − The sphere is quite large relative to the 'small' description
- − The plant lacks significant detail in the leaf textures
Stable Diffusion 3.5 Medium
- + Successfully placed the sphere inside the cube
- + Interesting floating effect of the sphere
- − The glass cube has inconsistent geometry and thickness
- − Texture of the red book is muddy and lacks clear edges
- − Overexposed lighting washes out the background details
Verdict: FLUX.2 [flex] produced a much higher quality image with realistic lighting, clean geometry, and professional composition. While Stable Diffusion 3.5 Medium followed the basic prompt instructions, it suffered from distorted glass edges and a lack of texture clarity compared to the polished output of FLUX.2.
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 Medium
AI Judge Analysis
FLUX.2 [flex]
- + Excellent skin texture and facial detail showing elder age naturally
- + Highly realistic mechanical interaction between the hands and the bicycle chain
- + Accurate depiction of motion blur on passing vehicles while maintaining a shallow depth of field
- − The framing feels slightly more polished than the requested 'imperfect framing'
Stable Diffusion 3.5 Medium
- + Successfully captured the 'imperfect framing' and candid street aesthetic
- + Nice atmospheric rendering of the rain and wet pavement reflections
- − Poor anatomical and mechanical detail in the man's hands and the bicycle structure
- − The man appears vertically compressed or hunched in a way that looks like a morphing artifact instead of natural posture
- − Lacks the skin texture and clarity requested in the prompt
Verdict: FLUX.2 [flex] produced a much higher quality image with superior skin textures, realistic mechanical details, and convincing depth of field. Stable Diffusion 3.5 Medium captured the 'candid' and 'imperfect' framing better, but failed significantly on the human anatomy and the intricate details of the bicycle.
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 Medium
AI Judge Analysis
FLUX.2 [flex]
- + Excellent depiction of ornate engraved armor with clear holy iconography
- + High-quality texture on the leather straps and knit cloth underlayer
- + Clean and professional-looking bokeh effects and lighting
- − The scars look slightly superficial and flat compared to the rest of the skin texture
Stable Diffusion 3.5 Medium
- + Intense and realistic skin textures including dirt and grime
- + Strong atmospheric lighting with a more dramatic color palette
- + Lifelike eye contact and expression
- − The detail on the armor engravings is significantly softer than Model A
- − Failed to include the requested beads in the braided hair
- − Some artifacts in the hair texture, making it look a bit unnatural in spots
Verdict: FLUX.2 [flex] produced a much more detailed and accurate rendering of the 'ornate engraved plate armor' and the specific requested elements like hair beads and cloth textures. While Stable Diffusion 3.5 Medium captured a more intense, grit-focused atmosphere, it missed several prompt details and lacked the sharpness found in FLUX.2.
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 Medium
AI Judge Analysis
FLUX.2 [flex]
- + Excellent typography with legible headers and clear price formatting.
- + High-quality, appetizing food photography that looks professional.
- + Strong adherence to the requested layout using colorful accents and a distinct grid.
- − The placeholder text (lorem ipsum style) is a bit messy in the descriptions.
- − Mismatched content where a bruschetta is listed under a 'Pizza' header.
Stable Diffusion 3.5 Medium
- + Good variety of food imagery in a consistent grid format.
- + Captures the 'minimalist' color palette well with lots of white space.
- − Text is largely unintelligible and poorly rendered.
- − Font choice is not the 'bold sans-serif' requested, appearing more like a stylized display font.
- − Layout feels cluttered and less professional than a real menu.
Verdict: FLUX.2 [flex] produced a much more realistic and usable menu design with clean typography and professional-grade food photography. While it had some minor content mismatches (like listing bruschetta as pizza), Stable Diffusion 3.5 Medium failed significantly on text legibility and grid organization.
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 Medium
AI Judge Analysis
FLUX.2 [flex]
- + Excellent typography with a cohesive fiery texture and glow.
- + Perfect execution of the 'exploded' burger concept with clearly suspended layers.
- + High textural detail on the bun, meat, and melting sauce.
- − The starburst design is slightly generic and overlaps the bottom bun.
Stable Diffusion 3.5 Medium
- + Good photorealistic rendering of the patty and lettuce textures.
- + Effective use of actual fire in the background for a dramatic effect.
- − Failed the 'exploded' instruction as the burger is mostly assembled.
- − Text rendering is plain and lacks the requested 'fiery effect'.
- − The starburst is a clip-art style illustration that clashes with the photographic style.
Verdict: FLUX.2 [flex] followed every part of the prompt, successfully creating a dynamic exploded burger with stylized, fiery typography that fits the ad theme. Stable Diffusion 3.5 Medium failed to separate the burger layers (the 'exploded' look) and used basic, non-stylized text that did not match the requested fiery aesthetic.
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 Medium
AI Judge Analysis
FLUX.2 [flex]
- + Excellent text rendering with perfect spelling of all requested items.
- + Authentic chalk texture and smudges that create a realistic chalk-on-board effect.
- + The handwriting shows high consistency while maintaining natural variations in stroke.
- − The date is placed on its own line rather than following the title as requested.
- − The handwriting style is slightly more 'chalk marker' than 'rough stick chalk'.
Stable Diffusion 3.5 Medium
- + Features a variety of chalk styles and flourish elements.
- + The wooden frame has a nice natural texture.
- − Failed significantly on text legibility, with numerous spelling errors and garbled words.
- − Ignored the specific date requested, showing 'APRIL $0624' instead.
- − Composition is messy with random lines and overlapping elements that make it look disorganized.
Verdict: FLUX.2 [flex] is the clear winner as it followed every instruction, including specific menu items and pricing, with near-perfect legibility and a beautiful aesthetic. Stable Diffusion 3.5 Medium struggled with text adherence, producing nonsensical words like 'Oocclutches' and failing to render the requested date correctly.
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 Medium
AI Judge Analysis
FLUX.2 [flex]
- + Excellent adherence to the 'horse on top' spatial instruction
- + High cinematic quality with vibrant lighting and a detailed nebula background
- + Creative anatomical interaction where the astronaut is holding the horse's legs like handles
- − The horse's hind legs are slightly anatomical confusing in how they flow into the body
Stable Diffusion 3.5 Medium
- + Realistic lighting on the astronaut's suit reflection
- + Clear, high-contrast background with the Earth below
- − Completely failed the negative constraint to have the horse on top
- − Noticeable anatomical artifacts in the horse's legs appearing as segmented tubes
- − Generic composition compared to the specific surreal request
Verdict: FLUX.2 [flex] followed the complex spatial instruction perfectly, creating a surreal and cinematic scene where the horse is riding the astronaut. Stable Diffusion 3.5 Medium failed the primary prompt constraint, instead generating a standard astronaut-on-horse image with significant anatomical errors in the horse's legs.
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 Medium
AI Judge Analysis
FLUX.2 [flex]
- + Excellent prompt adherence with the businesswoman looking at her phone as requested.
- + The capybara's anatomy is well-integrated into the human-like jacket and steering wheel placement.
- + Superior photographic realism and lighting consistency between the interior and the Manhattan background.
- − The steering wheel looks a bit small relative to the capybara's grip.
- − The texture on the capybara's hands appears more like human skin than fur or paws.
Stable Diffusion 3.5 Medium
- + Creative front-facing composition that highlights the capybara's expression.
- + Good rendering of the taxi driver cap with a detailed emblem.
- − Failed to include the phone in the businesswoman's hands, making her look out at the driver instead.
- − The anatomy of the paws on the steering wheel is anatomically incorrect and blurry.
- − The lighting on the capybara's face feels slightly flat and less photorealistic compared to the background.
Verdict: FLUX.2 [flex] successfully captured every element of the prompt, including the specific detail of the passsenger looking at her phone with a bored expression. Stable Diffusion 3.5 Medium opted for a front-facing perspective that missed the phone detail and struggled with the anatomy of the paws on the steering wheel. FLUX.2 [flex] is the clear winner for its superior realism and adherence to the narrative elements of the scene.
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 Medium
AI Judge Analysis
FLUX.2 [flex]
- + Excellent typography with perfect spelling and font choice.
- + Beautifully rendered central jack-o-lantern with realistic cinematic lighting.
- + Clean and professional graphic design composition with a detailed parchment border.
- − The parchment background is a bit desaturated compared to traditional 'vintage' paper.
Stable Diffusion 3.5 Medium
- + Includes multiple jack-o-lanterns and more prominent twisted trees.
- + Vibrant color palette with a clear night sky background.
- − Severe text rendering issues with misspellings like 'Halloweeen Inviloween'.
- − Visual artifacts such as a pumpkin floating in a tree without a neck.
- − The overall composition feels more like a child's storybook than a polished invitation.
Verdict: FLUX.2 [flex] produced a professional-grade invitation with perfect adherence to the text requirements and a sophisticated, cohesive aesthetic. In contrast, Stable Diffusion 3.5 Medium failed significantly on text rendering and produced a cluttered composition with several anatomical and logical errors in the artwork.
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 Medium
AI Judge Analysis
FLUX.2 [flex]
- + Perfectly followed text instructions including alignment and font weight.
- + Clean isometric 3D miniature aesthetic with a visible diorama base.
- + Higher diversity of sushi types and more accurate representation of the 'cartoon' style requested.
- − The flag icon is slightly large relative to the text.
- − The textures are a bit flat, leaning more into 'clay' than 'refined PBR'.
Stable Diffusion 3.5 Medium
- + Good glossy material finish on the roe and seaweed.
- + Solid color background is vibrant and clean.
- − Failed to include the requested flag icon.
- − Text rendering is messy with overlapping characters in 'SUSHI'.
- − Missing the 'raised diorama base' component specified in the prompt.
Verdict: FLUX.2 [flex] is the clear winner as it followed every prompt instruction, including the specific text layout, the flag icon, and the diorama base. Stable Diffusion 3.5 Medium failed on the text legibility and omitted several key visual components like the flag and the isometric platform.
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 Medium
AI Judge Analysis
FLUX.2 [flex]
- + Successfully included all four requested animals: puppy, kitten, bunny, and fox.
- + Excellent interpretation of atmospheric lighting with clear god rays and dew sparkles.
- + High level of anatomical realism and fur texture for all animals.
- − The butterflies appear somewhat repetitive in design and color.
Stable Diffusion 3.5 Medium
- + Strong, vibrant colors that create a joyful, warm aesthetic.
- + Good fur texture on the fox and puppy characters.
- − Failed to include the requested baby bunny animal.
- − The kitten has an unusual, almost fox-like head shape and ear size.
- − The lighting looks significantly more artificial and 'CGI' compared to the requested photorealism.
Verdict: FLUX.2 [flex] followed the prompt much more accurately, including all four specific animals while maintaining a highly photorealistic style. Stable Diffusion 3.5 Medium failed to include the bunny and struggled with the anatomical distinctness of the kitten, leaning into a more saturated, digital art style rather than the requested masterpiece realism.
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 Medium
AI Judge Analysis
FLUX.2 [flex]
- + Excellent typography including the correct use of the accent mark in 'Caffè'.
- + Perfect adherence to the minimalist vector emblem style.
- + Accurate rendering of the 'Est. 1720' banner and cloche icon.
- − The steam element is very faint and could be more pronounced.
Stable Diffusion 3.5 Medium
- + Intricate vintage illustration style with good use of texture.
- + Warm tones and shading evoke a classic feel.
- − Failed on typography, misspelling both 'Caffee' and 'Florrian'.
- − Incorrectly rendered the date as 'Est 170'.
- − The composition is cluttered and does not meet the 'minimalist' requirement.
Verdict: FLUX.2 [flex] followed every aspect of the prompt perfectly, producing a clean, professional vector logo with impeccable spelling and date accuracy. In contrast, Stable Diffusion 3.5 Medium failed significantly on the text elements—misspelling the name and the date—and ignored the 'minimalist' constraint in favor of a complex, cluttered illustration.
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 Medium
AI Judge Analysis
FLUX.2 [flex]
- + Excellent typography with perfect spelling and legibility.
- + Clean, professional flat-vector aesthetic that matches the requested NASA palette.
- + Accurate and logical flow of mission steps through clear iconography.
- − Missing the final 6th 'Landing' step from the prompt, stopping at Descent.
Stable Diffusion 3.5 Medium
- + Includes more steps along the vertical axis.
- + Follows the requested color palette reasonably well.
- − Terrible text rendering with many gibberish words like 'Lauan strip' and 'Trranllity'.
- − Chaotic composition with messy lines and confusing iconography.
- − Fails to maintain a clean, modern vector aesthetic, appearing more like a rough sketch.
Verdict: FLUX.2 [flex] produced a high-quality, professional-grade infographic with perfect text and clean vector art, clearly following the thematic instructions. In contrast, Stable Diffusion 3.5 Medium failed significantly on text legibility and graphic clarity, resulting in a cluttered and nonsensical layout. FLUX.2 is the clear winner despite missing the very last step of the sequence.
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