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]
#14 of 62 in Text-to-Image
Qwen Image 2512
#30 of 62 in Text-to-Image
Where the votes landed
FLUX.2 [flex]
55.6%
win rate
Ties
22.2%
Qwen Image 2512
22.2%
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.2 [flex]
- + Excellent adherence to spatial layout and lighting instructions.
- + High visual clarity and clean rendering of the cube and book.
- + Realistic depth of field with the plant correctly positioned behind the glass.
- − The sphere is quite large, pushing the definition of 'small' in the prompt.
Qwen Image 2512
- + Realistic texture on the book cover and wooden table.
- + The scale of the 'small' sphere is more accurate relative to the cube.
- − Confusing glass physics where the back panels look more like mirrors than transparent glass.
- − The plant is significantly less visible through the glass compared to Model A.
Verdict: FLUX.2 [flex] produced a much clearer and more logically coherent image, where the plant is distinctly visible through the transparent glass cube as requested. Qwen Image 2512 struggled with the transparency of the cube, making the internal surfaces look mirrored and obscuring the plant behind it.
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.2 [flex]
- + Excellent adherence to the motion blur and candid photography request
- + Realistic skin textures and clothing details
- + Effective use of shallow depth of field and bokeh
- − The anatomy of the bicycle is slightly nonsensical with the rear wheel and pedal arrangement
- − The man is kneeling directly on the wet pavement which looks a bit unnatural
Qwen Image 2512
- + Very realistic facial features and skin texture
- + Strong composition with a focus on the subject's expression
- + Better bicycle anatomy compared to the competitor
- − Missed the 'motion blur from passing cars' instruction; the cars in the background are stagnant
- − The man is posing for the camera rather than 'repairing' the bike as requested
Verdict: FLUX.2 [flex] followed the prompt's technical requirements much better, successfully incorporating the motion blur and the action of repairing the bike, though the bike's structure is flawed. Qwen Image 2512 produced a high-quality portrait, but the subject is stationary and posing, ignoring the specific request for a candid repair scene with motion-blurred cars. FLUX.2 [flex] is the winner for its superior prompt adherence and atmospheric storytelling.
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.2 [flex]
- + Extremely crisp and detailed plate armor engraving.
- + Realistic skin texture and well-defined scars.
- + Perfect adherence to 'small beads' in the hair braids.
- − Lighting feels slightly artificial and flat compared to the other image.
- − The background fire looks a bit like a stock asset.
Qwen Image 2512
- + Exceptional atmosphere with very realistic torchlight and shadows.
- + Skin texture looks more 'battle-worn' and weathered.
- + Excellent depth of field with cinematic spark bokeh.
- − Ornate engraving on the gorget is slightly less sharp than Model A.
- − The beads in the hair are a bit more generic.
Verdict: Both models followed the prompt exceptionally well, but Qwen Image 2512 wins due to its superior lighting and cinematic atmosphere, which better captures the 'battle-worn' mood. FLUX.2 [flex] has slightly sharper technical details on the armor, but the overall composition feels more like a studio portrait than a scene from a fantasy world.
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.2 [flex]
- + Excellent text rendering with clear, legible English words.
- + Clean, professional typography using the requested bold sans-serif fonts.
- + Consistent and high-quality food photography that fits the grid layout.
- − The 'Appetizers' section contains photos but no corresponding text menu items.
- − Repeats nearly identical pizza/steak images in the grid.
Qwen Image 2512
- + Dynamic grid layout with a good variety of colorful food images.
- + Bold use of color accents and icons to denote different sections.
- + Includes a larger focal image at the bottom for visual interest.
- − Poor text rendering with many gibberish words and spelling errors.
- − The typography feels cluttered and less professional compared to a real menu.
- − Price formatting is inconsistent and unrealistic.
Verdict: FLUX.2 [flex] produced a much more usable and professional design with legible text and a clean minimalist aesthetic. While Qwen Image 2512 had an interesting layout with more color, its failure to render coherent English text or realistic pricing makes it less effective as a design asset.
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.2 [flex]
- + Excellent typography with a cohesive fiery texture across all text elements.
- + High photorealistic detail on the ingredients, especially the freshness of the lettuce and juicy patty.
- + Strong adherence to the 'starburst' pricing element with a dynamic, glowing finish.
- − The composition of the burger ingredients feels slightly stiff compared to the 'exploded' request.
Qwen Image 2512
- + More dynamic 'exploded' composition with ingredients tilting and fragments flying more naturally.
- + Captures the sense of motion well with flying crumbs and sauce droplets.
- + Includes additional requested details like the sauce in a visually appealing way.
- − Failed to include the word 'TIME' in the 'LIMITED TIME ONLY' message.
- − Ingredients like the tomato and onion look slightly more rendered/artificial than in Image A.
Verdict: FLUX.2 [flex] produced a more polished advertisement with perfect text rendering and superior photorealism of the food itself. While Qwen Image 2512 captured the 'exploded' motion more dynamically, it missed a word in the required text prompt and the starburst element was less impactful. FLUX.2 [flex] is the winner for its professional finish and exact adherence to all text requirements.
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.2 [flex]
- + Perfect spelling of all menu items including complex words like 'Risotto'.
- + Realistic chalk smudging and background texture that feels like a used chalkboard.
- + Followed all text instructions including the price and the final cookie item completely.
- − The handwriting style is a bit uniform and lacks some of the 'elegant cursive' flair requested for the title.
Qwen Image 2512
- + Very beautiful and elegant cursive handwriting for both the title and menu items.
- + Excellent chalk texture on the individual letters with varying thickness.
- − Contains a spelling error: 'Risitto' instead of 'Risotto'.
- − The layout of the third item is a bit disjointed with the price floating far to the right.
Verdict: FLUX.2 [flex] wins because it managed perfect spelling and a more cohesive layout, capturing the specific gritty reality of a chalkboard. Qwen Image 2512 produced more beautiful handwriting but suffered from a spelling error ('Risitto') and less natural spacing for the prices.
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.2 [flex]
- + Excellent adherence to the 'horse on top' spatial instruction
- + Cinematic lighting and high level of detail in the background galaxy
- + Creative interpretation of the horse physically riding the astronaut
- − Anatomical issues with the horse's front legs merging into the astronaut's hands
- − The horse's back legs appear slightly disconnected
Qwen Image 2512
- + High photographic realism and sharp textures
- + Clean rendering of the astronaut suit
- − Failed the primary negative constraint: the astronaut is riding the horse
- − The harness straps on the horse's face are physically incoherent
- − Lacks the 'surreal' quality requested in the prompt
Verdict: FLUX.2 [flex] successfully followed the difficult spatial instruction to place the horse on top of the astronaut, creating a truly surreal and cinematic image. Qwen Image 2512 completely ignored the specific 'not vice versa' instruction, producing a standard astronaut-on-horse image which was explicitly discouraged.
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.2 [flex]
- + Excellent texture on the capybara's fur and the jacket fabric.
- + Cinematic lighting with a clear New York City street blur visible through the window.
- + Composition clearly shows the passenger's expression and activity concurrently with the driver.
- − The capybara's hands/paws look slightly too human-like and distorted on the steering wheel.
- − The scale of the capybara relative to the car interior feels a bit large.
Qwen Image 2512
- + Perfectly captures the 'bored' expression of the passenger as requested.
- + The frontal composition creates a symmetrical and engaging humorous contrast.
- + Includes a realistic seatbelt detail on the capybara.
- − The capybara's paws look like leather gloves or claws rather than natural anatomy.
- − The perspective through the windshield is somewhat generic and lacks the depth of the city shown in the other model.
Verdict: Both models followed the prompt instructions excellently, capturing the surreal scenario with high photorealism. FLUX.2 [flex] excels in environmental lighting and textured details, making for a more atmospheric image, while Qwen Image 2512 does a superior job capturing the specific 'bored' facial expression of the passenger for better comedic effect. FLUX.2 [flex] is the slight winner for overall composition and visual polish.
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.2 [flex]
- + Perfect text accuracy for all requested fields
- + Authentic vintage parchment aesthetic with torn edges
- + Clean, professional layout that matches the gothic theme
- − The parchment borders are a bit repetitive with the skulls in every corner
- − Background trees are a bit sparse compared to the foreground detail
Qwen Image 2512
- + Vibrant cinematic lighting and rich colors
- + Excellent integration of the thorns and spiderwebs into the border
- + Highly detailed twisted trees and atmospheric background
- − Spelling error in the main title ('Hallowern' instead of 'Halloween')
- − The text layout feels slightly more cramped than Model A
Verdict: FLUX.2 [flex] successfully followed all text instructions with 100% accuracy and captured the vintage parchment look perfectly. While Qwen Image 2512 has more impressive cinematic lighting and intricate illustrative details, the spelling error in the main title ('Hallowern') makes it less functional as a real invitation.
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.2 [flex]
- + Perfect text rendering and alignment.
- + Ultra-clean minimalist aesthetic exactly matching the prompt.
- + High-quality, soft 3D cartoon textures.
Qwen Image 2512
- + Intricate diorama base with organic details like grass and leaves.
- + Playful custom typography style.
- + High level of detail on the sushi pieces themselves.
- − The flag icon is skewed and placed awkwardly next to the text.
- − Text is not centered perfectly as requested.
- − The diorama base has slight texture inconsistencies for a 'clean' look.
Verdict: FLUX.2 [flex] produced an exceptionally clean and balanced image that adhered perfectly to every technical instruction, including centering and text placement. Qwen Image 2512 offered more creative flair in the diorama base but struggled with the specific layout and icon placement requested in the prompt.
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.2 [flex]
- + Excellent depiction of movement and playfulness as requested
- + Includes all four specified animals with distinct, proportional sizes
- + Beautiful environmental lighting with clear god rays and atmospheric depth
- − The fox's front right leg has a slightly unnatural anatomy
- − Some butterflies appear to be floating stamps without much depth
Qwen Image 2512
- + Very high detail in fur texture and facial expressions
- + Strong character interaction with animals huddling together
- + Accurate butterflies and sharp foreground rendering
- − Fails to capture the 'playfully chasing' and 'tumbling' aspect of the prompt
- − The composition feels a bit cramped and posed like a studio portrait
- − The golden retriever's head is disproportionately large compared to the other animals
Verdict: FLUX.2 [flex] is the winner because it successfully captures the dynamic action of the prompt, showing the animals chasing butterflies and tumbling in a wide meadow. While Qwen Image 2512 has slightly sharper textures, it ignored the 'chasing' and 'tumbling' instructions in favor of a static, posed group portrait where the scale of the animals feels inconsistent.
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.2 [flex]
- + Perfect adherence to the 'minimalist' and 'vector emblem' descriptors.
- + Clean, professional execution of the typography and icon.
- + Extremely accurate rendering of text and the requested banner element.
- − Steam effect is very subtle and simple compared to the other elements.
- − The banner styling is slightly generic.
Qwen Image 2512
- + Excellent artistic detail and shading on the cloche.
- + Beautiful, dynamic steam interpretation.
- + Strong vintage aesthetic with high-quality cross-hatching textures.
- − Fails the 'minimalist' requirement with its complex shading.
- − Typography is a bit heavy, and the letter 'a' in 'Florian' has some slight structural weirdness.
- − Large, ornate steam clouds might be too busy for a standard logo.
Verdict: FLUX.2 [flex] perfectly captures the 'minimalist vector emblem' request, resulting in a logo that is clean, professional, and ready for use. Qwen Image 2512 produces a much more detailed and visually striking illustration with beautiful textures, but it ignores the 'minimalist' constraint and is less successful as a practical logo design.
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.2 [flex]
- + Excellent typography with zero spelling errors.
- + Clean, professional flat-vector aesthetic that perfectly matches the 'modern infographic' request.
- + Logical layout that follows the mission progression clearly.
- − Missed the final 'Landing' step, stopping at descent.
- − The Saturn V rocket illustration is slightly skewed.
Qwen Image 2512
- + Included all 6 requested steps including the landing on the surface.
- + Effective use of the NASA-inspired color palette.
- + Good representative icons for the Lunar Module.
- − Significant spelling errors and gibberish text (e.g., 'Translaurtcoit', 'Desceeint').
- − Messy layout with overlapping text and repeated step numbers.
- − Included prompt instructions ('Steps stop at landing:') as literal text in the image.
Verdict: FLUX.2 [flex] produced a vastly superior graphic design with perfect text rendering and a high-end vector feel, though it missed the final step of the mission. Qwen Image 2512 followed the step-by-step instructions more literally but failed significantly on text quality, layout organization, and followed the prompt's meta-instructions as literal text, resulting in a cluttered and unprofessional appearance.
Explore each model
Improved version of Alibaba's Qwen image model with better text rendering, finer natural textures, and more realistic human generation.