Distilled version of Black Forest Labs' FLUX.2 [dev] outperforming it at a cheaper price. Developed by fal.ai.
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX.2 [dev] Turbo
#3 of 62 in Text-to-Image
Qwen Image 2512
#30 of 62 in Text-to-Image
Where the votes landed
FLUX.2 [dev] Turbo
54.5%
win rate
Ties
18.2%
Qwen Image 2512
27.3%
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 [dev] Turbo
- + Excellent photographic realism with natural lighting and depth of field.
- + Accurate glass physics, showing convincing reflections and refractions of the sphere and plant.
- + The plant is clearly behind the cube as requested, seen through the glass.
- − The glass cube has an open top rather than being a solid enclosure, though the book still sits on the edges.
Qwen Image 2512
- + Perfectly follows all spatial instructions, including the sphere being 'inside' a closed-top cube.
- + Clean composition with vibrant colors.
- − The reflections on the glass are logically inconsistent and messy, particularly the extra sphere appearing on the right.
- − Lower visual realism compared to Model A, appearing more like a 3D render.
Verdict: FLUX.2 [dev] Turbo produces a much more realistic image with sophisticated lighting and material properties, though the cube is missing its top face. Qwen Image 2512 follows the geometry of the prompt more literally but suffers from significant glass reflection artifacts and a less natural aesthetic.
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 [dev] Turbo
- + Excellent adherence to the 'repairing' action with tools on the ground.
- + Strong execution of light rain and detailed reflections on the wet pavement.
- + Realistic skin textures and natural, 'imperfect' candid framing.
Qwen Image 2512
- + Very realistic facial features and skin texture.
- + Good use of motion blur in the background cars.
- + Captures the cinematic street photography aesthetic well.
- − The man is posing for the camera rather than 'repairing' the bicycle as requested.
- − The bicycle structure is warped, particularly the front wheel and fork area.
- − The rain effect is very subtle, nearly missing the atmosphere requested.
Verdict: FLUX.2 [dev] Turbo followed the prompt much more accurately, showing the man actively engaged in repairing the bike with tools spread on the ground. Qwen Image 2512 produces a high-quality portrait, but the subject is just posing next to the bike, and the bike itself has significant anatomical distortions.
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 [dev] Turbo
- + Excellent depiction of ornate engraving on the plate armor
- + Highly detailed realism in the eyes and skin texture
- + Clear rendering of leather straps with visible buckles
- − The torch in the background is slightly blurry and lacks some internal flame detail
Qwen Image 2512
- + Strong atmospheric lighting with warm glows reflecting off the metal
- + Good integration of the braided hair and beads into the character design
- + Effective use of shallow depth of field for a cinematic feel
- − The engraving on the armor is less sharp and detailed compared to the competitor
- − The facial scars look a bit more like surface marks than depth-based wounds
Verdict: Both models followed the prompt exceptionally well, capturing the battle-worn paladin aesthetic accurately. FLUX.2 [dev] Turbo is the winner due to its superior clarity and micro-detail, particularly in the armor engravings and the texture of the leather straps, whereas Qwen Image 2512 has a slightly more painterly, soft look in some of the metalwork.
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 [dev] Turbo
- + Excellent text legibility and font choice
- + Professional and clean grid layout
- + High-quality, realistic food photography
- − Nonsense filler text in the subtitle and descriptions
- − Section headers are slightly disjointed (Mains is separated from the others)
Qwen Image 2512
- + Stronger use of vibrant color accents in the UI
- + Clearer three-section structure for Appetizers/Mains/Pizzas
- + Good grid of food photos
- − Very poor text rendering and spelling (RESSAGRENT, APPETIIZIZERS)
- − Unrealistic pricing and currency values
- − Lower image clarity and some visual artifacts in the food photos
Verdict: FLUX.2 [dev] Turbo produces a far more professional and usable design with high-quality photography and legible, albeit nonsensical, text. Qwen Image 2512 follows the requested structural sections more closely but suffers from severe spelling errors and a less polished aesthetic.
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 [dev] Turbo
- + Excellent typography with a consistent fiery glow effect across all text elements.
- + Highly photorealistic food textures, especially on the bun and tomatoes.
- + Clean and professional composition suitable for a high-end food advertisement.
- − The 'exploded' effect is a bit subtle compared to the potential for dynamic motion.
Qwen Image 2512
- + Stronger sense of 'exploded' motion with small debris and particles flying outwards.
- + Bold, high-impact background and lighting that matches the fiery theme well.
- − Missed part of the requested text, omitting the word 'TIME' from 'LIMITED TIME ONLY'.
- − The text rendering on the main title is slightly less integrated with the fire effect than in Model A.
Verdict: FLUX.2 [dev] Turbo followed all text instructions perfectly, including the full phrase 'LIMITED TIME ONLY', and provided superior photorealism in the food rendering. While Qwen Image 2512 captured a more dynamic sense of an explosion, it failed on prompt adherence by missing a word in the secondary message.
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 [dev] Turbo
- + Natural chalk texture with realistic dust and smudges.
- + Handwriting has excellent character and variation.
- + Successfully completed the truncated prompt 'Brown But...' as 'Brown Butter Chocolate Chip Cookies'.
- − Has a slight formatting error where a stray dollar sign appears before jumping to the new line for the price of the first item.
- − The date text is slightly less elegant than Model B's.
Qwen Image 2512
- + Excellent typography style with high consistency and legibility.
- + Clean layout with very professional-looking cursive elements.
- + Followed the truncated prompt instructions perfectly.
- − Misspelled 'Risotto' as 'Risitto'.
- − The chalk texture looks a bit too 'clean' or digital compared to the grainy texture of Model A.
Verdict: Both models handled the text rendering and the completion of the truncated prompt exceptionally well. FLUX.2 [dev] Turbo has a more authentic chalk-on-blackboard texture with realistic imperfections, while Qwen Image 2512 produces a cleaner, more stylized look but fails on the spelling of 'Risotto'.
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 [dev] Turbo
- + Excellent cinematic lighting and complex background with planets and dust.
- + Dynamic composition with a sense of motion in the horse's mane and tail.
- + Highly detailed spacesuit and textures.
- − Failed the specific spatial instruction; the astronaut is on top of the horse.
Qwen Image 2512
- + Features a clear, well-lit subject with good anatomical consistency on the horse.
- + High resolution with sharp details on the spacesuit.
- − Failed the specific spatial instruction; the astronaut is on top of the horse.
- − The background is relatively simple compared to Model A.
Verdict: Both FLUX.2 [dev] Turbo and Qwen Image 2512 failed the negative/inverse constraint of placing the 'horse on top' of the astronaut, instead defaulting to the common trope of an astronaut riding a horse. FLUX.2 [dev] Turbo is preferred because it delivers a much more cinematic and atmospheric image with superior lighting and environmental details.
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 [dev] Turbo
- + Excellent photorealism in the capybara's fur and clothing textures.
- + Superior lighting consistency with logical reflections on the dashboard.
- + Accurate representation of the businesswoman looking down at her phone.
- − The passenger is placed in the front seat instead of the back seat as requested.
- − The capybara's paws lack detail and look slightly like human-animal hybrid hands.
Qwen Image 2512
- + Correctly places the businesswoman in the back seat as specified in the prompt.
- + Excellent captured 'bored' expression on the human passenger.
- + Strong composition that centers the driver and provides clear depth to the interior.
- − Textural quality on the capybara's face is slightly less realistic than Model A.
- − The steering wheel and hands look a bit rubbery and lacked fine detail.
Verdict: While FLUX.2 [dev] Turbo offers slightly better skin and fur textures, it failed the spatial requirement of having the passenger in the back seat. Qwen Image 2512 followed all instructions accurately, including the background placement and the specific 'bored' expression, making it the better overall representation 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.”
AI Judge Analysis
FLUX.2 [dev] Turbo
- + Perfect text rendering with no spelling errors
- + Excellent atmospheric composition with a cohesive vintage parchment texture
- + Strong adherence to all prompt elements including the specific border details
- − The bats are slightly clunky in shape compared to the rest of the illustration
Qwen Image 2512
- + High visual clarity and vibrant cinematic lighting on the jack-o-lantern
- + Elegant scroll banner design with artistic curls
- + Dynamic and spooky forest background
- − Spelling error in the main title ('Hallowern' instead of 'Halloween')
- − The thorny border is less integrated with the parchment theme compared to Model A
Verdict: FLUX.2 [dev] Turbo followed the prompt instructions perfectly, particularly regarding text accuracy and the specific parchment aesthetic. Qwen Image 2512 had slightly more vibrant lighting and a more intricate scroll, but it failed on a critical detail by misspelling the word 'Halloween' in the main header.
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 [dev] Turbo
- + Excellent rendering of textures, particularly the grain of the wood and the translucency of the fish.
- + Perfectly clean and legible typography that follows the prompt exactly.
- + Very high visual clarity and professional lighting that enhances the 3D diorama effect.
- − A bit conservative with the 'miniature' aesthetic, feeling more like a realistic render than a cartoon-style model.
Qwen Image 2512
- + Strong adherence to the 'cartoon scene' and 'miniature' style with rounded, soft-shaded 3D assets.
- + The typography is stylized and fits the aesthetic of the scene.
- + Good variety of sushi types on the diorama.
- − The text border has slight artifacts and is not as clean as Model A.
- − The 'Japan' text is slightly off-center compared to the 'Sushi' text.
- − The flag icon is slightly distorted.
Verdict: Both models followed the prompt exceptionally well, but FLUX.2 [dev] Turbo takes the lead with its superior rendering of materials and perfectly crisp typography. While Qwen Image 2512 captured the 'cartoon' aesthetic slightly better, FLUX.2 presented a much cleaner and more professional-looking image that felt more like high-quality 3D art.
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 [dev] Turbo
- + Dynamic composition with animals playfully tumbling and chasing butterflies.
- + Strong adherence to lighting requests, including dew sparkles and subtle god rays.
- + Excellent anatomical accuracy for all four distinct animals.
Qwen Image 2512
- + High level of fur detail and clear textures.
- + Strong, dramatic god rays and warm golden lighting.
- + Charming, expressive faces on all animals.
- − Static, posed composition fails to capture the 'chasing' and 'tumbling' action requested.
- − Anatomical blending issues where the puppy's paws appear to be the rabbit's or kitten's paws.
- − The fox kit looks slightly more like a mature fox than a baby kit.
Verdict: FLUX.2 [dev] Turbo followed the prompt much more accurately by depicting the animals in motion (tumbling and chasing) rather than a static pose. While Qwen Image 2512 has beautiful lighting and texture, it suffers from several anatomical merging issues where the limbs of the animals overlap in a confusing, AI-artifacted way.
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 [dev] Turbo
- + Perfect adherence to the 'minimalist' and 'vector emblem' style.
- + Clean, professional typography that is highly readable.
- + Excellent use of negative space and balanced composition.
- − The steam illustration is a bit generic compared to the rest of the logo.
Qwen Image 2512
- + Highly detailed illustrative style with great metallic shading on the cloche.
- + Dynamic steam effects that add movement to the image.
- − Misses the 'minimalist' requirement, leaning more towards a complex illustration.
- − The banner and text are slightly crowded and less balanced than Model A.
Verdict: FLUX.2 [dev] Turbo followed the prompt instructions more accurately, providing a truly minimalist vector-style logo with clean typography and a well-proportioned banner. While Qwen Image 2512 created a visually impressive and detailed illustration, it ignored the 'minimalist' keyword, resulting in a design that is too complex for a standard logo.
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 [dev] Turbo
- + Excellent typography with nearly perfect spelling
- + Followed the instructions for specific iconography and steps accurately
- + Clean, high-quality vector illustrations for the lunar module and rocket
- − Confused layout with redundant icons (two Earths, multiple translunar labels)
- − The Saturn V label includes the literal word 'icon' from the prompt
Qwen Image 2512
- + Stronger vertical 'poster' composition that flows better visually
- + Accurate NASA-inspired color palette and flat-vector style
- + Successfully numbered the stages in a logical sequence
- − Significant spelling errors in technical labels (e.g., 'TranslauraJ', 'Desceeint')
- − The numbering is inconsistent, repeating '2' and '3' twice
Verdict: FLUX.2 [dev] Turbo produced much higher quality text and more detailed icon assets, though its spatial arrangement was a bit cluttered and repetitive. Qwen Image 2512 captured the intended poster layout better but suffered from poor spelling and illogical numbering of the mission steps. FLUX.2 [dev] Turbo is the winner for its clarity and accuracy in following the specific content of the prompt.
Explore each model
Improved version of Alibaba's Qwen image model with better text rendering, finer natural textures, and more realistic human generation.