Black Forest Labs' aesthetically-tuned 12-billion parameter flow transformer optimized for high-quality images with incredible aesthetics, suitable for personal and commercial use
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX.1 Krea [dev]
#47 of 62 in Text-to-Image
Z-Image Turbo
#12 of 62 in Text-to-Image
Where the votes landed
FLUX.1 Krea [dev]
0.0%
win rate
Ties
0.0%
Z-Image Turbo
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 Krea [dev]
- + Excellent depiction of lighting and reflections on the sphere and glass surfaces.
- + Highly realistic textures on the red book and the wooden table.
- + Perfect adherence to spatial prompts, including the plant being visible through the glass.
- − The sphere is quite large relative to the cube, pushing the 'small' descriptor in the prompt.
Z-Image Turbo
- + Accurately depicts the sphere as 'small' relative to the cube.
- + Good placement of all required elements.
- + Crisp lighting and clear lens depth of field.
- − The glass cube has strange white artifacts/splotches on the right side.
- − The bottom of the cube has some geometric inconsistencies where it meets the table.
- − Texture on the table and plant is slightly less natural than Model A.
Verdict: FLUX.1 Krea [dev] is the winner due to its superior rendering of light, shadows, and reflections, which are crucial for a scene involving glass. While Z-Image Turbo captured the scale of the 'small' sphere better, it suffered from visual artifacts and less convincing material textures compared to the highly realistic output of FLUX.1 Krea [dev].
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 Krea [dev]
- + Excellent depiction of motion blur in the background cars as requested.
- + Effective cinematic depth of field with realistic wet pavement reflections.
- + Captures a convincing 'candid' street photography aesthetic.
- − The bike anatomy is slightly warped, particularly where the frame meets the rear wheel.
- − The man's posture is more standing than actively 'repairing'.
Z-Image Turbo
- + Natural skin texture and realistic arm anatomy for an elderly man.
- + Excellent rendering of raindrops on the bicycle frame and mudguard.
- + The bicycle design is structurally more coherent and realistic.
- − Fails to include the requested motion blur on passing cars, which appear frozen.
- − The lighting on the pavement is less cinematic and more flat compared to the other model.
- − The framing feels very centered and intentional, missing the 'imperfect framing' prompt.
Verdict: FLUX.1 Krea [dev] followed the technical camera prompts much better, providing the requested motion blur and cinematic lighting that gives the image a professional photography feel. While Z-Image Turbo has impressive fine detail on the subject's skin and the water droplets, it failed to execute the motion blur and lighting mood requested in the prompt.
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 Krea [dev]
- + Strictly follows the 'close portrait' instruction with a symmetrical, impactful headshot.
- + Superior engraving detail on the plate armor with very high resolution.
- + Excellent representation of bokeh sparks in the background to create depth.
- − The lighting feels a bit more studio-like and less like a flickering torch compared to the competitor.
- − The blood/dirt on the face looks slightly painted on rather than integrated into the skin texture.
Z-Image Turbo
- + Excellent depiction of the torchlight source providing realistic warm highlights across the metal.
- + Great diversity in textures, showing the chainmail and quilted underlayers clearly.
- + The messy braiding and beads feel more natural and battle-worn.
- − Less of a 'close portrait' than requested, pulling back to a medium shot.
- − The metal engraving is slightly less crisp and defined than in the other image.
- − The torch flame looks a bit flat and stylistically disconnected from the rest of the scene.
Verdict: FLUX.1 Krea [dev] followed the framing instructions better, providing a true close-up portrait with stunningly detailed engraving on the armor. However, Z-Image Turbo captured the spirit of the 'torchlight' prompt much more effectively by including a visible light source that casts realistic highlights, even though it failed to provide as close a crop as requested. FLUX.1 Krea [dev] is the winner for its superior texture quality and adherence to composition constraints.
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 Krea [dev]
- + Excellent structure that mimics a real physical menu page.
- + Food photography is well-integrated and looks high-quality.
- + The text layout (items, descriptions, and prices) is highly realistic.
- − Text consists of gibberish characters and symbols rather than legible English.
- − Includes a duplicate section header for 'Appetizers'.
Z-Image Turbo
- + Text is largely legible English and uses very bold, modern sans-serif fonts.
- + High-contrast orange accents provide a strong visual identity.
- + Composition is balanced and easy to read quickly.
- − 'Pizza Mans' contains a clear typo.
- − The grid layout feels a bit repetitive with the large food photos across the top.
- − Section labeling is slightly confusing with 'Settion' written vertically.
Verdict: FLUX.1 Krea [dev] produces a more convincing overall menu mock-up with professional spacing and appetizing photography, but fails on text legibility. Z-Image Turbo produces a more usable design with bold, readable English text and vibrant accents, though it suffers from some typos and a slightly less sophisticated layout. Z-Image Turbo is preferred for its legible content and adherence to the 'bold font' and 'vibrant' requirements.
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 Krea [dev]
- + Excellent typography rendering with clean, professional fonts.
- + High-quality photorealistic texture on the bun and patties.
- + Includes all requested text elements accurately.
- − The price is not in a starburst as requested.
- − The 'exploded' effect is minimal, with most components still touching.
- − Text does not have the requested fiery/glowing effect.
Z-Image Turbo
- + Perfect adherence to the fiery/glowing text effect for all titles.
- + Successfully placed the price in a glowing starburst shape.
- + Very dynamic lighting and embers that match the 'fiery' theme.
- − The burger is not exploded/deconstructed; it is a solid stacked burger.
- − Contains stray artifacts/gibberish around the bottom edges.
Verdict: FLUX.1 Krea (dev) produces a cleaner, more professional-looking advertisement with superior text legibility and realistic textures, though it fails on the specific styling of the price tag. Z-Image Turbo captures the 'fiery' aesthetic and starburst requirement much better, but it completely misses the 'exploded' burger request. FLUX.1 Krea (dev) is the likely winner for its overall image quality and better interpretation of a deconstructed burger.
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 Krea [dev]
- + Excellent chalk texture including dust smears and smudges.
- + Elegant cursive style matches the aesthetic requested.
- + High photorealism in the lighting and frame texture.
- − Several spelling errors and hallucinated words (e.g., 'Gruffle', 'Lcoman', 'Chvcuations').
- − Text rendering is slightly inconsistent in quality toward the bottom.
- − Failed to precisely list only the three requested items including the partial description.
Z-Image Turbo
- + Highly legible and clean text rendering.
- + Accurate spelling for almost all items, despite a minor typo in 'Mustroom'.
- + Followed the list of items more accurately than the competitor.
- − The 'chalk' looks more like a digital font and lacks the authentic grainy texture requested.
- − The handwriting looks too uniform and lacks natural variation.
- − Composition is a bit clinical and lacks the 'cozy café' lighting seen in the other image.
Verdict: FLUX.1 Krea (dev) produces a much more realistic image with convincing chalk textures and atmospheric lighting, but it fails significantly on text legibility and accuracy. Z-Image Turbo provides much clearer text and better adherence to the specific menu items, but the text looks like a digital overlay rather than realistic chalk. FLUX.1 Krea (dev) is the likely winner for its superior artistic quality and texture, which were key parts of the prompt, despite the spelling mistakes.
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 Krea [dev]
- + Excellent cinematic lighting and high-quality textures.
- + Creative interpretation of the horse integrated with space technology.
- + Superior background with Earth and realistic star field.
- − Failed the specific spatial instruction for the horse to be on top of the astronaut.
Z-Image Turbo
- + Clear, sharp image with realistic horse and space suit textures.
- + Good composition with active posing of the horse.
- − Failed the specific spatial instruction for the horse to be on top of the astronaut.
- − The space background is less convincing and lacks cinematic depth compared to Image A.
- − The horse's front legs exhibit anatomical clipping/merging.
Verdict: Both models failed to follow the specific 'horse on top, not vice versa' instruction, which usually indicates an astronaut carrying a horse. However, FLUX.1 Krea [dev] produced a much more cinematic and visually coherent image with superior lighting and a more detailed environmental background. Z-Image Turbo's output feels like a standard composite and has minor 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.”
AI Judge Analysis
FLUX.1 Krea [dev]
- + Follows the specific request for the woman being in the back seat.
- + Captures the 'inside the taxi' perspective more accurately with a wide shot.
- + Correctly renders the taxi light on top of the car roof.
- − Includes an extra woman in the front passenger seat that was not requested.
- − The 'hat' looks more like a plastic toy or a hard hat rather than a driver cap.
- − The capybara's hands look more like paws but the anatomy is slightly awkward.
Z-Image Turbo
- + Excellent rendering of the capybara's fur and the leather texture of the seat.
- + The driver cap is highly realistic and well-integrated.
- + The lighting and skin tones on the passenger are very high quality.
- − Failed to place the passenger in the back seat; she is in the front passenger seat.
- − The capybara's hand on the steering wheel looks like a human hand with fur rather than a capybara paw.
- − The composition is a bit tight, missing the 'interior' feel of the whole taxi.
Verdict: While Z-Image Turbo produces a much more photorealistic and high-fidelity image with superior textures, it fails a key part of the prompt by placing the passenger in the front seat. FLUX.1 Krea [dev] captures the requested scene layout better by putting the passenger in the back, but it adds an extra person and has less realistic textures. Overall, Z-Image Turbo is preferred for its striking visual quality despite the spatial error.
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 Krea [dev]
- + Excellent ornamental border design incorporating webs and thorns
- + Moody cinematic lighting on the jack-o-lantern
- + Clean, high-quality digital illustration style
- − Significant spelling errors in the title and body text
- − Layout is a bit segmented with overlapping scrolls
Z-Image Turbo
- + Successfully captured the 'dark parchment' texture requested
- + Text is largely legible and follows the prompt's layout
- + Strong atmosphere with the moody night sky and twisted trees
- − Small typo in 'Archves' for the location
- − The thorns and webs on the outer border look a bit messy and detached
Verdict: Z-Image Turbo is the winner as it accurately followed the text instructions and technical requirements for a parchment poster. While FLUX.1 Krea [dev] had a more polished border, it failed the primary task by misspelling 'Party' as 'Pasty' and distorting much of the secondary text.
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 Krea [dev]
- + Perfect adherence to the requested flag (Japanese flag)
- + High-quality realistic PBR materials and lighting
- + Clean, accurate font rendering and placement
- − The salmon texture looks slightly more like a render than a 'cartoon' style
Z-Image Turbo
- + Very pleasing soft cartoon aesthetic
- + Clean isometric diorama base layout
- + Good text legibility
- − Displays the Chinese flag instead of the requested Japanese flag
- − Low-tier resolution or upscaling artifacts visible on the rice grains
Verdict: FLUX.1 Krea (dev) followed the prompt perfectly, including the specific cultural context of the flag. Z-Image Turbo failed a key instruction by generating a Chinese flag for a Japanese-themed scene and has significantly lower detail quality on the sushi textures.
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 Krea [dev]
- + Excellent fur detail and individual hair rendering
- + Strong dynamic composition with animals in motion
- + Beautiful golden hour lighting with clear god rays
- − Failed to include the baby bunny
- − Included two kittens instead of one
- − Anatomical issues with the leaping kitten's paws
Z-Image Turbo
- + Correctly included all requested animals: puppy, kitten, bunny, and fox
- + Very expressive facial features on all animals
- + Includes visible dew sparkles as requested
- − The puppy's paw is awkwardly merging with the bunny's back
- − Lower overall resolution and slightly softer fur textures compared to Model A
- − Composition feels a bit more static and crowded
Verdict: Z-Image Turbo is the winner for prompt adherence as it successfully included all four specific animals, whereas FLUX.1 Krea [dev] missed the bunny and duplicated the kitten. While FLUX.1 Krea [dev] produced higher technical quality in terms of fur texture and lighting, Z-Image Turbo's ability to follow complex subject lists makes it more reliable for this specific prompt.
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 Krea [dev]
- + Excellent vintage engraving style with detailed cross-hatching
- + Captures the 'vintage' and 'subtle texture' requirements perfectly
- + Artistic and high-quality illustration of the cloche and steam
- − Spelling error in the main brand name ('Flanor'in' instead of 'Florian')
- − Typography is cluttered and hard to read
Z-Image Turbo
- + Perfect spelling of 'Caffè Florian'
- + Clean, minimalist vector aesthetic that fits a modern-minimalist logo style
- + Highly legible typography and well-balanced composition
- − Missed the 'banner' requirement for the 'Est. 1720' text
- − The texture is very faint, appearing almost as a flat cream background
Verdict: While FLUX.1 Krea (dev) has a much richer artistic style and better interpretation of the 'vintage' and 'banner' prompts, it fails significantly on text legibility and spelling. Z-Image Turbo captures the minimalist vector style and renders the text perfectly, making it a more practical choice for a logo despite missing some specific design elements like the banner.
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 Krea [dev]
- + Successfully captures a cohesive space/poster aesthetic.
- + Follows the numbered step structure requested in the prompt.
- + Strong adherence to the specified navy, white, and red color palette.
- − The icons do not match the specific mission steps (e.g., Saturn with rings for step 1, generic people for step 4).
- − Text rendering is messy with several gibberish labels.
- − The flow of the infographic is confusing and lacks logical sequence.
Z-Image Turbo
- + Icons are much more relevant to the mission, including a rocket and lunar module.
- + Clean, modern flat-vector UI design style.
- + Text is largely legible and identifies correct mission phases like 'Earth Orbit'.
- − Misspells 'Apollo' as 'Apolio' in the main header.
- − Fails to include all six specific numbered steps requested.
- − The 'Saturn V' rocket icon looks more like a generic stylized shuttle/rocket.
Verdict: Z-Image Turbo is the winner because its iconography actually relates to the moon mission (Lunar Module, Earth, Moon) whereas FLUX.1 Krea produced generic icons like Saturn and human figures that didn't fit the steps. While Z-Image Turbo failed to number all six steps and had a typo in the title, its visual quality and relevance to the 'modern vector' request were significantly higher.
Explore each model
Tongyi-MAI's 6-billion parameter distilled text-to-image model optimized for speed, achieving high-quality generation in 8 steps or fewer with support for bilingual text rendering