Black Forest Labs' 12-billion parameter flow transformer for high-quality text-to-image generation, suitable for personal and commercial use with streaming support
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX.1 [dev]
#16 of 62 in Text-to-Image
FLUX.1 [schnell]
#48 of 62 in Text-to-Image
Where the votes landed
FLUX.1 [dev]
0.0%
win rate
Ties
0.0%
FLUX.1 [schnell]
100.0%
win rate
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.”
AI Judge Analysis
FLUX.1 [dev]
- + Perfect adherence to object placement instructions
- + Highly realistic textures on the leather-bound red book
- + Convincing soft lighting and reflections on the glass
- − The glass structure is more like a hollow frame than a solid cube
- − A secondary ghost-like reflection of the sphere appears on the back pane
FLUX.1 [schnell]
- + Clean glass rendering with logical internal reflections
- + Good lighting consistency with the window source
- + High image clarity and sharpness
- − Failed the spatial prompt by placing a second sphere on top of the book
- − The plant isn't logically visible 'through' the glass as requested due to the perspective and cube thickness
Verdict: FLUX.1 [dev] followed the complex spatial instructions perfectly, placing objects exactly where requested. FLUX.1 [schnell] hallucinated an additional sphere on top of the book, which violates the prompt requirements.
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.”
AI Judge Analysis
FLUX.1 [dev]
- + Excellent shallow depth of field with realistic background bokeh
- + Compelling natural skin texture and facial expression
- + Superior rendering of light rain and atmospheric moisture
- − The man is holding the handlebars rather than performing a repair
- − The bicycle design is a bit modern and basic for a 'candid' street scene
FLUX.1 [schnell]
- + Strong composition with a more detailed red bicycle
- + Better adherence to the 'repairing' aspect of the prompt
- + Captures a distinct Japanese street aesthetic with signage and crosswalk
- − Fails to include the requested motion blur on passing cars
- − Hands are poorly rendered with anatomical artifacts
- − The man's hair and clothing textures look slightly more artificial
Verdict: FLUX.1 [dev] produces a much more realistic and cinematic image with superior lighting and texture, though it misses the specific 'repair' action. FLUX.1 [schnell] captures the repair activity and street environment better, but suffers from significant hand artifacts and fails to implement the requested motion blur. FLUX.1 [dev] is the winner for its high visual fidelity and convincing 'no stylization' look.
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.”
AI Judge Analysis
FLUX.1 [dev]
- + Excellent depiction of ornate engraved plate armor with realistic metal textures.
- + Accurately represents 'bokeh sparks' in the background.
- + Strong facial focus with lifelike eyes and subtle skin texture.
- − The character looks relatively clean and youthful despite the 'battle-worn' and 'dirt' prompt.
- − Missing the beads in the braids specified in the prompt.
FLUX.1 [schnell]
- + Followed the detail about beads in the hair braids and braided leather accents.
- + Strong 'battle-worn' aesthetic with intense expression and visible scars.
- + Superior texture on the leather and cloth underlayer.
- − The lighting is very harsh, creating an almost painterly rather than photographic look.
- − The armor engraving is less 'ornate' than requested compared to the other model.
Verdict: FLUX.1 [dev] produced a more visually appealing and photorealistic image with beautiful metalwork, but it missed the specific detail about beads and felt less 'battle-worn'. FLUX.1 [schnell] captures the grit, the beads, and the requested textures much better, making it the superior choice for prompt adherence despite the slightly over-processed lighting.
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.”
AI Judge Analysis
FLUX.1 [dev]
- + Excellent typography with clean, legible heading fonts.
- + Professional use of white space and subtle curved graphic elements.
- + High-quality, isolated food photography that fits a minimalist aesthetic.
- − Failed to include exactly four images as a grid, opting for floating food photos instead.
- − The text at the bottom is nonsensical.
FLUX.1 [schnell]
- + Successfully implemented the requested food photo grid layout.
- + Included all specific sections: Appetizers, Pizza, and Mains.
- + Strong adherence to the minimalist, white background requirement.
- − Text rendering is very poor and looks like scribbles rather than realistic font characters.
- − The grid images have lower quality and less appealing lighting compared to the food in the other model.
Verdict: FLUX.1 [dev] produces a much more professional-looking graphic design with superior typography and high-end food photography, though it missed the specific grid layout request. FLUX.1 [schnell] followed the structural instructions (grid and specific sections) more accurately but failed significantly on visual quality and text legibility, resulting in a less usable design.
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.”
AI Judge Analysis
FLUX.1 [dev]
- + Excellent photographic rendering of the burger ingredients
- + Clean and centered composition following the 'exploded' vertical style
- + Accurate text rendering for the secondary messages
- − Completely missed the 'MAGIC BURGER' title text
- − Failed to include the starburst element for the price
- − The background is simplistic with flat fire at the bottom rather than an immersive fiery environment
FLUX.1 [schnell]
- + Dynamic 'shatter' explosion effect with high sense of motion
- + Included the starburst graphic as requested
- + Background and lighting are more dramatic and consistent with the fiery theme
- − Typos in text including 'AGIC BURGER' instead of 'MAGIC BURGER' and a '€699' error in the starburst
- − The 'exploded' burger is mostly intact with random bits flying around rather than a decomposed stack
- − Lower visual clarity on the secondary falling ingredients
Verdict: FLUX.1 [dev] produced a much higher quality image of an actual burger with clear, appetizing textures, but it failed to include the primary title text and starburst. FLUX.1 [schnell] followed more components of the prompt including the starburst and primary heading, but it suffered from significant spelling errors and lost the 'exploded stack' aesthetic for a more chaotic debris effect. FLUX.1 [dev] is the winner for superior image quality and lack of distractions, despite the missing text.
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.”
AI Judge Analysis
FLUX.1 [dev]
- + Excellent text rendering with near-perfect spelling and legibility.
- + The chalk texture is highly realistic with dust-like artifacts and varying pressure.
- + Great artistic composition that looks like a real cafe board.
- − Slightly struggled with the cursive requirement for the title, opted for a clean hand-lettered print instead.
- − Small spelling error at the bottom with 'drish' instead of 'fresh'.
FLUX.1 [schnell]
- + Natural handwriting slant and variations in letter size.
- + Cozy cafe background elements are visible in the reflection and background.
- − Major spelling errors across almost all menu items including 'Taffle Mushmnctiomn'.
- − Poor layout with text being clipped at the edges or oddly spaced.
- − Repetitive lines and nonsensical words toward the bottom of the board.
Verdict: FLUX.1 [dev] is the clear winner as it successfully rendered almost all the requested text with high legibility and a realistic chalk aesthetic. FLUX.1 [schnell] failed significantly on prompt adherence, producing garbled text and incorrect menu items despite the handwritten style looking authentic.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
AI Judge Analysis
FLUX.1 [dev]
- + High visual quality and cinematic lighting
- + Excellent rendering of the astronaut suit and horse hair
- − Failed the spatial reasoning instruction, placing the astronaut on top
FLUX.1 [schnell]
- + Successfully followed the difficult spatial instruction of horse on top
- + Captures a surreal, dreamlike atmosphere
- + Creative interpretation of the requested inverse hierarchy
- − Major anatomical anomalies including multiple horse heads
- − Coherence issues where the horse and astronaut merge
Verdict: This challenge highlights a trade-off between prompt adherence and anatomical logic. While FLUX.1 [dev] produced a much cleaner and more professional image, it completely ignored the specific instruction to have the horse on top; FLUX.1 [schnell] followed the difficult instruction but suffered from significant glitches like a two-headed horse. Because FLUX.1 [schnell] actually attempted the surreal spatial request while the other did not, it is the more relevant response despite the artifacts.
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.”
AI Judge Analysis
FLUX.1 [dev]
- + Excellent capybara paws placement on the steering wheel.
- + The passenger is correctly located in the back seat as requested.
- + High level of fur detail and realistic lighting on the capybara.
- − The passenger's scale seems slightly off, appearing too small relative to the capybara.
- − The camera perspective from outside the windshield makes the car layout look a bit confusing.
FLUX.1 [schnell]
- + Natural composition from inside the car looking at the driver and passenger.
- + The capybara's professional driver uniform and jacket are very well-rendered.
- + The passenger's bored expression is captured effectively.
- − The passenger appears to be sitting in the front passenger seat rather than the back seat.
- − One of the capybara's paws is not on the steering wheel, failing the prompt's specific instruction.
- − The steering wheel is positioned too far to the left, almost outside of the car frame.
Verdict: FLUX.1 [dev] followed the spatial instructions much better, correctly placing the passenger in the back seat and ensuring both paws were on the wheel. While FLUX.1 [schnell] had a more pleasing interior composition, it failed several specific prompt details regarding character placement and actions.
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.”
AI Judge Analysis
FLUX.1 [dev]
- + Strong composition with a clear focal point.
- + Excellent thorn border detail that frames the central artwork well.
- + Highly polished, clean vector-style illustration.
- − Significant spelling errors in the title text.
- − Redundant text elements like '7pm, 7pm' in the details section.
- − Vertical orientation despite the 'square format' instruction.
FLUX.1 [schnell]
- + Successfully followed the square format instruction.
- + Clear gothic title text with large, legible lettering.
- + Evocative color palette with nice atmospheric fog effects.
- − Messy typography with numerous typos and repeated words.
- − Lower quality thorn border compared to the competitor.
- − The text layout feels clustered and poorly aligned.
Verdict: Both models struggled significantly with the specific text requirements, though FLUX.1 [dev] produced a much more visually cohesive and professionally illustrated central image. FLUX.1 [schnell] followed the square format instruction better, but the text rendering was chaotic and the overall composition felt unbalanced compared to the clean, vintage aesthetic of FLUX.1 [dev].
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.”
AI Judge Analysis
FLUX.1 [dev]
- + Excellent miniature plastic/clay aesthetic consistent with the 'cartoon scene' prompt
- + Good use of the diorama base and decorative elements like the flowers
- + Accurate 45-degree isometric composition
- − Failed the text prompt, displaying 'SUSH CATON' instead of 'SUSHI'
- − Includes hallucinated kanji-like characters at the top
FLUX.1 [schnell]
- + Perfect adheres to the text requirement for 'JAPAN'
- + Very clean, minimal, and high-clarity rendering
- + Accurate isometric perspective and solid background adherence
- − Completely omitted the secondary text 'SUSHI'
- − The single piece of sushi looks slightly less 'cartoon miniature' compared to Model A
Verdict: Both models followed the isometric and light blue background instructions well. FLUX.1 [dev] captured the 'miniature cartoon' atmosphere much better with its soft textures and framing, but it failed significantly on the text rendering. FLUX.1 [schnell] produced a cleaner, more professional graphic but missed half of the requested text and felt slightly more generic.
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.”
AI Judge Analysis
FLUX.1 [dev]
- + Excellent distribution of light and soft bokeh
- + Captures a very joyful and whimsical mood
- − Failed to include a tabby kitten, providing two dogs instead
- − The art style is more stylized/cartoonish than the requested hyper-photorealistic
- − Anatomical issues with the paws of the fox/bunny hybrids
FLUX.1 [schnell]
- + Successfully included all requested animal types including the kitten and fox
- + Much more realistic fur textures and detail
- + Dynamic composition with more interesting interactions between the animals
- − The fox kit has coloring that leans slightly too orange/saturated
- − Foreground flowers are a bit blurred, though this is likely intentional for depth
Verdict: FLUX.1 [schnell] is the clear winner as it adhered to all specific animals in the prompt, including the tabby kitten and fox kit which FLUX.1 [dev] failed to distinguish. Additionally, FLUX.1 [schnell] achieved a much more realistic level of detail and lighting, whereas FLUX.1 [dev] produced a 2D-looking, stylized aesthetic that missed the 'hyper-photorealistic' instruction.
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.”
AI Judge Analysis
FLUX.1 [dev]
- + Strong minimalist vector aesthetic
- + Includes clear steam element from the cloche
- + Successfully includes 'Est. 1720' text
- − Misspelled the primary name as 'Flarilaan'
- − Misspelled 'Restaurant' as 'Reseaurant'
- − Cloche relative size is a bit small compared to typography
FLUX.1 [schnell]
- + Clean vector emblem layout
- + Good use of the warm brown and cream color palette
- + Excellent center-aligned composition
- − Significant text errors including 'Framilan' and 'Est. 7720'
- − Missing the steam element requested in the prompt
- − Cloche appears more like a building dome than food service hardware
Verdict: Both models failed to correctly spell the name 'Caffè Florian', but FLUX.1 [dev] produced a much more accurate date and included the specific requested detail of steam rising from the cloche. FLUX.1 [schnell] creates a cleaner layout but hallucinates the year as '7720' and fails to include the steam detail.
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.”
AI Judge Analysis
FLUX.1 [dev]
- + Excellent adherence to the requested color palette.
- + Includes distinct iconography at the bottom for lunar, descent, and landing phases.
- + Superior infographic composition with clear labels and a path of travel.
- − Text consists of gibberish despite the header being clear.
- − Includes irrelevant planets with rings like Saturn which were not requested.
FLUX.1 [schnell]
- + Features a very clean, symmetrical layout.
- + Follows the flat-vector style requirements well.
- + Displays the descent and landing phase visually on a moon-like surface.
- − Fails to include the specific six-step progression requested in the prompt.
- − The rocket design is clunky and does not accurately resemble the Saturn V.
- − Text is mostly garbled and repetitive.
Verdict: FLUX.1 [dev] is the clear winner as it more accurately captures the 'infographic' essence of the prompt, including a multi-step flow and a bottom legend for specific icons. While FLUX.1 [schnell] creates a clean image, it fails to represent the requested chronological mission steps, providing a more abstract and less informative layout.
Explore each model
Black Forest Labs' 12 billion parameter distilled image generation model optimized for speed, capable of generating high-quality images in just 4 inference steps