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]
#50 of 62 in Text-to-Image
Qwen Image 2512
#30 of 62 in Text-to-Image
Where the votes landed
FLUX1.1 [pro]
0%
win rate
Ties
0%
Qwen Image 2512
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
FLUX1.1 [pro]
- + Excellent photorealistic rendering of materials, especially the glass edges and reflections.
- + Accurate lighting direction from the left with realistic light catch on the sphere.
- + Higher resolution and finer texture on the red book cover.
- − The glass object is a rectangular prism rather than a perfect 'cube'.
- − The sphere appears to be floating unnaturally in the center rather than resting on the base.
Qwen Image 2512
- + Correctly interprets the 'cube' geometry more accurately than the rival model.
- + Realistic reflections of the sphere on the base of the glass box.
- + Strong source preservation of the plant being visible through the glass panels.
- − The glass has a slight cyan/teal tint which wasn't requested.
- − The lighting is a bit flatter and less dynamic than the other model.
Verdict: Both models followed the prompt instructions well, including the specific placement of the plant and red book. FLUX1.1 [pro] produced a more visually stunning and photorealistic image with better textures, though it failed to make the glass container a true cube; Qwen Image 2512 followed the geometric instruction of a 'cube' more accurately while maintaining a very naturalistic scene.
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
FLUX1.1 [pro]
- + Excellent bokeh and photographic lighting
- + The bicycle design looks specifically Japanese (mamachari style)
- + Captures a candid 'action' feel with the subject leaning over the bike
- − Major structural failure with the rear half of the bicycle and the man's legs
- − The bicycle frame is fragmented and missing the rear wheel
- − Subject's hands are mangled and blending into the handlebars
Qwen Image 2512
- + Natural and highly detailed skin texture on the subject's face
- + Much better anatomical and structural coherence for both the man and the bicycle
- + Successfully captures the specific requested 'light rain' visual effect
- − The subject is looking directly at the camera, which feels less like a 'candid' shot
- − Slightly less 'motion blur' on the background cars than requested
- − Missing some of the imperfect 'street photography' framing requested
Verdict: While FLUX1.1 [pro] captures a more cinematic atmosphere and better bokeh, it suffers from severe anatomical and structural hallucinations where the bicycle and the man's lower body essentially disappear or mesh together. Qwen Image 2512 provides a much more coherent and realistic image with impressive skin textures and a fully rendered bicycle, despite the subject making eye contact with the camera.
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
FLUX1.1 [pro]
- + Exceptional realism in the skin texture and lifelike eyes.
- + Very natural shallow depth of field with organic bokeh.
- + High-quality lighting integration across the face and hair.
- − Missed the request for beads in the hair braids.
- − Ornate armor engraving is mostly cropped out of the frame.
Qwen Image 2512
- + Perfect adherence to specific details like hair beads and leather straps.
- + Exquisite intricate engraving on the plate armor.
- + Clear visualization of the requested 'battle-worn' state with distinct scars.
- − Skin texture looks slightly more digital/painterly compared to the other model.
- − The background flame is a bit distracting and less out-of-focus than requested.
Verdict: FLUX1.1 [pro] produces a more photorealistic portrait with stunning eye detail, but it misses several specific prompt instructions such as the hair beads. Qwen Image 2512 followed every part of the prompt meticulously, including the ornate armor, leather straps, and beaded braids, making it the better interpretation despite slightly less realistic skin Rendering.
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
FLUX1.1 [pro]
- + Excellent professional presentation including realistic silverware for context
- + Typography is cleaner and follows standard menu hierarchy better
- + The photo grid of the food is integrated naturally into the center page
- − Text consists of garbled placeholder-like characters
- − The 'Pizza' photo section is slightly cluttered with overlapping graphic boxes
Qwen Image 2512
- + Strong adherence to the 'grid' request with a clear 4x2 top layout
- + Bold use of color accents effectively distinguishes categories
- + Higher quality rendering of the food photography itself
- − Significant spelling errors in prominent headings like 'RESSAGRENT' and 'APPESTIIZERS'
- − The layout feels slightly cramped with too many photos at the top
Verdict: FLUX1.1 [pro] provides a more realistic and professional mock-up of a menu, looking ready for a restaurant website, though its text is less legible. Qwen Image 2512 followed the specific 'grid' prompt more literal and used bolder accents, but suffered from noticeable spelling errors and a slightly more cluttered composition. FLUX1.1 [pro] is the winner for its better balance of minimalist design and overall presentation quality.
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
FLUX1.1 [pro]
- + Excellent photographic quality and lighting on the burger textures.
- + Clean, modern typography that is perfectly legible.
- + Detailed particles and debris that add a sense of impact.
- − The burger is not truly 'exploded' with all components suspended; it is mostly assembled.
- − Missed the 'starburst' and 'fiery glowing effect' requirements for the text.
- − Redundant text appearing three times.
Qwen Image 2512
- + Stronger adherence to the 'exploded' instruction with separated layers.
- + Excellent execution of the fiery, glowing text effect and the starburst element.
- + Dynamic composition with a sense of verticality and motion.
- − Missed the word 'TIME' in 'LIMITED TIME ONLY'.
- − The bun texture is slightly less realistic compared to Model A.
- − Some small artifacts in the flying sauces and embers.
Verdict: Qwen Image 2512 is the clear winner for following the specific layout instructions, providing the requested 'exploded' burger view and the fiery starburst element. While FLUX1.1 [pro] has slightly superior photorealism on the burger patty itself, it failed to separate the ingredients in mid-air and ignored the stylistic prompts for the 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
FLUX1.1 [pro]
- + Excellent background atmosphere and lighting
- + Captures the casual café aesthetic well
- − Hallucinates and repeats lines, turning 'Grilled Octopus' into two separate entries with incorrect prices
- − The handwriting looks more like a digital marker font than actual chalk texture
- − Spelling errors in 'Chipkies', 'Herbss', and 'giluten'
Qwen Image 2512
- + Authentic chalk texture with realistic smudges and dust on the board
- + Highly accurate text rendering of the complex menu items and prices in a beautiful script
- + Consistently follows the date and specific pricing requested in the prompt
- − Minor spelling error in 'Risitto' (Risotto)
- − The layout is a bit cramped at the bottom edge
Verdict: Qwen Image 2512 is the clear winner for its superior text rendering and authentic chalk texture, whereas FLUX1.1 [pro] fails to follow the menu items correctly, repeating lines and hallucinating incorrect prices like $228. Qwen Image 2512 also captures the requested 'natural variation' and 'elegant cursive' much more successfully than the digital-looking script in FLUX.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
AI Judge Analysis
FLUX1.1 [pro]
- + Excellent cinematic lighting and atmosphere
- + Strong artistic rendering of the horse and clouds
- + High level of detail in the astronaut suit
- − Completely failed the negative constraint to have the horse on top
- − The horse features anatomical weirdness where its chest merges with the astronaut's seat
Qwen Image 2512
- + Clear, high-resolution rendering of the astronaut's face
- + Good lighting consistency with the Earth's curvature
- + Clean textures on the horse's coat
- − Completely failed the specific spatial instruction for the horse to be on top
- − The horse's front legs exhibit strange anatomical positioning and hovering
Verdict: Both FLUX1.1 [pro] and Qwen Image 2512 failed the most difficult part of the prompt, which was the 'horse on top' spatial instruction. FLUX1.1 [pro] is preferred as it captures the 'cinematic' and 'surreal' style requested much better than the flatter, more generic composition of Qwen Image 2512.
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
FLUX1.1 [pro]
- + Excellent photorealistic lighting and bokeh effects.
- + Accurate high-quality textures on the capybara's fur and the woman's coat.
- − Failed the core composition by placing the woman in the passenger seat instead of the back seat.
- − The capybara is not holding the steering wheel and appears to be sitting sideways.
Qwen Image 2512
- + Followed all compositional instructions, including placing the passenger in the back seat and paws on the steering wheel.
- + The expression of the human passenger perfectly matches the requested 'bored' look.
- + The capybara's professional uniform cap is well-rendered and fitting for the scene.
- − The capybara's 'front paws' look slightly more like human fingers in gloves than animal paws.
- − Slightly less artistic bokeh and cinematic lighting compared to the competitor.
Verdict: Qwen Image 2512 is the clear winner because it accurately followed the complex spatial instructions, placing the passenger in the back seat and ensuring the capybara's paws were on the steering wheel. FLUX1.1 [pro] failed several prompt requirements, placing the passenger in the front seat and missing the steering wheel interaction entirely, despite having superior lighting and texture quality.
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
FLUX1.1 [pro]
- + Strong cinematic lighting with a moody color palette
- + Atmospheric background composition with the moon and trees
- − Several typos in the text including 'inivted' and 'frightss'
- − Redundant duplication of the event details at the bottom
- − Failed the square aspect ratio requirement
Qwen Image 2512
- + Accurately followed the square format request
- + Perfect spelling on almost all text elements including the small banner
- + Excellent composition with the requested cobweb and thorn border
- − Minor spelling error in the word 'Hallowern' in the title
Verdict: Qwen Image 2512 is the superior choice for this task as it followed the square format instruction and successfully integrated all requested design elements like cobwebs and thorns. While FLUX1.1 [pro] has beautiful lighting, it suffered from numerous text repetitions and spelling errors that make the invitation unusable without significant editing.
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
FLUX1.1 [pro]
- + Excellent typography style that matches the 3D aesthetic
- + Very clean, soft PBR textures that look like high-quality 3D renders
- + Bright and balanced colors that enhance the miniature feel
- − Missed the flag icon mentioned in the prompt
- − Shadows on the base are slightly soft/diffuse compared to a rigid isometric look
Qwen Image 2512
- + Successfully included the Japanese flag icon as requested
- + Strong adherence to the 'small raised diorama base' with layered construction
- + High level of detail in the sushi textures, specifically the rice grains
- − Text rendering has some minor aesthetic inconsistencies in the outlining
- − Composition feels a bit more cluttered compared to the 'minimal' request
Verdict: Both models performed exceptionally well on the isometric layout and text rendering. Qwen Image 2512 followed the instructions more closely by including the flag icon and a distinct diorama base, whereas FLUX1.1 [pro] produced a cleaner, more aesthetically pleasing set of materials that felt more professional despite missing the icon.
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
FLUX1.1 [pro]
- + Expert use of backlighting and soft focus to create a magical atmosphere
- + Excellent horizontal composition with a unified upward gaze toward the butterflies
- − Failed to include the red fox kit (only three animals present)
- − The butterfly shapes are simplified and glowing rather than realistic
Qwen Image 2512
- + Successfully included all four requested animals (dog, cat, bunny, and fox)
- + High-fidelity rendering of fur textures and detailed monarch-style butterflies
- + Clear implementation of 'god rays' from the sunrise
- − The composition feels slightly crowded and posed like a studio portrait
- − Relatively static poses compared to the requested 'chasing' and 'tumbling'
Verdict: Qwen Image 2512 wins primarily on prompt adherence, as it managed to include all four requested animals (golden retriever, tabby kitten, bunny, and red fox), whereas FLUX1.1 [pro] missed the fox. While FLUX1.1 [pro] has a more artistic, cinematic lighting style, Qwen Image 2512 offers better detail in the butterflies and more literally follows the complex list of subjects.
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
FLUX1.1 [pro]
- + Excellent vintage border and engraving style texture.
- + Strong vector emblem composition that feels authentic to mid-century design.
- − Serious spelling error in the primary brand name ('Fratilian' instead of 'Florian').
- − The cloche is less recognizable and resembles a building dome.
Qwen Image 2512
- + Perfect adherence to text requirements with correct spelling.
- + Dynamic and clear illustration of the cloche and steam elements.
- + Excellent use of the requested banner for the date.
- − Less 'minimalist' than the prompt requested due to thick shadows and complex steam.
- − Missing the decorative border elements seen in the other version.
Verdict: Qwen Image 2512 is the clear winner because it correctly spells the brand name 'Caffè Florian', whereas FLUX1.1 [pro] fails to render the primary text correctly. Qwen Image 2512 also captures the metallic look of the cloche much more effectively while maintaining the requested vintage aesthetic.
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
FLUX1.1 [pro]
- + Excellent adherence to the 'flat-vector' style requested
- + Adheres well to the specific NASA-inspired color palette
- + Clean, minimalist composition suitable for a modern poster
- − Poor execution of the requested steps (duplicate numbers and disjointed flow)
- − Text is mostly gibberish or misspelt
- − Iconography is inconsistent with the specific stages listed (e.g., Earth icon used for Descent)
Qwen Image 2512
- + Successfully included all 6 requested steps with relevant illustrations
- + High-quality vector illustrations of the Saturn V and Lunar Module
- + Legible text and logical flow for an infographic
- − Included numbering errors and some spelling mistakes (e.g., 'Translaurtcit')
- − Less adherence to the 'flat' style, using more detailed shading and gradients than requested
- − Added elements not requested from the specific palette like orange/gold tones
Verdict: Qwen Image 2512 is the clear winner for its superior ability to follow the complex sequential instructions of the prompt, providing distinct and accurate illustrations for the Saturn V and Lunar Module stages. While FLUX1.1 [pro] captured the aesthetic 'flat vector' look better, it failed significantly to illustrate the actual mission steps or provide a logical infographic flow.
Explore each model
Improved version of Alibaba's Qwen image model with better text rendering, finer natural textures, and more realistic human generation.