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 17 shared challenges, with an AI judge weighing in on each.
FLUX.2 [flex]
#14 of 62 in Text-to-Image
Z-Image Turbo
#12 of 62 in Text-to-Image
Where the votes landed
FLUX.2 [flex]
42.9%
win rate
Ties
14.3%
Z-Image Turbo
42.9%
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 photographic quality and sharp focus
- + Accurately represents glass refraction and reflections
- + Clean, modern composition with well-balanced lighting
- − The blue sphere is quite large, whereas the prompt asked for a small one
Z-Image Turbo
- + Followed the 'small blue sphere' instruction more accurately
- + Realistic texture on the red book cover
- + Included all required elements in the scene
- − The background plant is very blurry and less distinct
- − The glass cube has strange reflective artifacts on the side panels
- − The bottom of the cube looks like a mirror rather than clear glass
Verdict: Both models followed the prompt perfectly in terms of spatial relationships and objects. FLUX.2 [flex] produced a much higher quality image with superior clarity and realistic glass physics, though the sphere was larger than requested. Z-Image Turbo followed the scale instruction for the sphere better but suffered from a muddy background and inconsistent rendering of the glass cube's base.
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 aesthetic prompts like 'cinematic', 'shallow depth of field', and 'reflections on wet pavement'.
- + Shows the subject actively 'repairing' the bicycle as requested.
- + Strong atmospheric lighting and convincing motion blur on background traffic.
- − The structural geometry of the bicycle frame is slightly nonsensical near the pedals.
- − The left foot appears to be merging into the ground/bike structure.
Z-Image Turbo
- + Realistic skin textures and clothing folds.
- + The person looks naturally Japanese and the environment feels authentic.
- + Captures the light rain effect well on the ground.
- − Fails the 'repairing' prompt; the man is simply standing with or pushing the bike.
- − Lacks the requested 'motion blur' from passing cars; the background car is static.
- − Missing the 'shallow depth of field' and 'cinematic' lighting requested.
Verdict: FLUX.2 [flex] is the clear winner as it adhered to almost every specific technical prompt, including motion blur, shallow depth of field, and the specific action of repairing the bike. Z-Image Turbo produced a high-quality, realistic image, but failed to capture the kinetic energy and cinematic atmosphere requested by the user, and ignored the primary action of 'repairing'.
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]
- + Excellent intricate engraving on the plate armor
- + Highly realistic skin texture with clear dirt and scar details
- + Perfect representation of small beads within the braided hair
- − The sparks look like static dots rather than motion-blurred bokeh embers
- − Lighting is a bit flat compared to the dramatic shadows of Image B
Z-Image Turbo
- + Very atmospheric local lighting from the torch reflecting on face and metal
- + Superior depth of field and bokeh spark effects
- + Excellent macro-level detail on the underlayer chainmail and cloth
- − The braids are a bit messy and less defined than requested
- − Face is slightly thinner and less symmetrical, though still realistic
Verdict: Both models followed the prompt exceptionally well, but FLUX.2 [flex] produced a cleaner, more detailed close-up with superior texture on the armor engravings. Z-Image Turbo captures the cinematic mood and lighting better, particularly with the warm glow and bokeh sparks, but it is slightly less focused as a 'close portrait'.
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]
- + Strictly followed the category requirements for appetizers, pizza, and mains.
- + Excellent layout balance with a clear hierarchy and clean white space.
- + Food photos are high quality and consistent with the category headers above or below them.
- − Text is mostly gibberish, though the letterforms are clean.
- − The placement of 'Appetizers' at the top with 'Pizza' and 'Mains' at the bottom feels slightly disconnected.
Z-Image Turbo
- + Bold, impactful typography that catches the eye immediately.
- + Generates recognizable currency symbols and numbers for pricing.
- + Vibrant food photography that fills the grid well.
- − Spelling errors in large headers such as 'MANS' and 'SETIIION'.
- − The layout is a bit cluttered, with food photos interrupting the flow between text sections.
- − Inaccurate categorization; food photos don't always align with the nearby text sections.
Verdict: FLUX.2 [flex] produced a much more professional and realistic menu layout that adheres perfectly to the requested sections (Appetizers, Pizza, Mains). While Z-Image Turbo has bolder text, its significant spelling errors and disjointed layout make it less functional as a design template compared to the clean, minimalist execution of FLUX.2 [flex].
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 adherence to the 'exploded' request with clear separation of ingredients.
- + Perfectly renders the starburst element and fiery text effects.
- + High photorealism in the textures of the meat patty and bun.
- − The sauce dripping from the cheese looks slightly unnatural in its thickness.
Z-Image Turbo
- + Clean, professional graphic design for the logo and secondary text.
- + Good color vibrance and lighting on the double-patty burger.
- + Accurate text rendering for all requested strings.
- − Fails to produce an 'exploded' burger, keeping most ingredients stacked.
- − The starburst is more of a sticker icon than a dynamic fiery explosion.
Verdict: FLUX.2 [flex] achieved a much better interpretation of the 'exploded' prompt, creating a dynamic sense of motion with suspended ingredients that looks like a high-end food advertisement. While Z-Image Turbo produced clean text and a delicious-looking burger, it failed to separate the components as requested and had a less impactful background.
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]
- + Excellent typography with a realistic chalk texture and smear effect.
- + Perfect spelling on all menu items including the final dessert.
- + Balanced composition with warm ambient lighting that fits the 'cozy café' prompt.
- − The 's' in 'Today's' is lowercase, which is a minor stylistic quirk.
Z-Image Turbo
- + Captures the requested content accurately including price points.
- + Good vertical alignment of menu items.
- − Contains a spelling error: 'Mustroom' instead of 'Mushroom'.
- − The title is in print-style block lettering rather than the requested 'elegant cursive chalk handwriting'.
- − The chalk texture looks a bit too clean and digital compared to Image A.
Verdict: FLUX.2 [flex] is the clear winner as it followed the handwriting style instructions perfectly and maintained perfect spelling throughout. Z-Image Turbo failed the style requirement for cursive handwriting and included a spelling typo in the first menu item.
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]
- + Successfully followed the difficult spatial instruction of the horse on top.
- + High visual quality with beautiful cinematic lighting and a vibrant space background.
- + Clearer rendering of surfaces, textures, and anatomical details.
- − The interaction between the horse's hooves and the astronaut's hands is slightly awkward.
Z-Image Turbo
- + Good clarity on the astronaut's suit details.
- − Failed the primary spatial prompt by placing the astronaut on top of the horse.
- − Flat, uninspired background that lacks the requested 'cinematic' feel.
- − Noticeable anatomy issues with the horse's back legs.
Verdict: FLUX.2 [flex] successfully interpreted the complex 'horse on top' prompt, which is a common failure point for most models, and delivered a highly detailed, cinematic image. Z-Image Turbo followed the standard 'astronaut riding horse' trope, ignoring the negative constraint, and produced a much lower quality composition with anatomical flaws.
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 lighting and atmosphere that captures the vibrant night scene of Manhattan.
- + Superior sharpness and textural detail on the capybara's fur and the taxi interior.
- + Strong prompt adherence with the capybara's calm expression and the woman's bored look.
- − The steering wheel placement looks slightly awkward relative to the capybara's hands.
- − The capybara's paws look a bit too human-like in their grasping structure.
Z-Image Turbo
- + Good composition with a clear view of both the driver and the passenger.
- + Successfully depicts the capybara wearing all requested clothing items.
- + Effective use of depth of field to keep the focus on the subjects.
- − The overall lighting is a bit flat and doesn't quite capture the 'night in NYC' feel as well as Model A.
- − Lower resolution and softer details compared to the competing image.
- − The city background is very generic and lacks the iconic Manhattan light blur.
Verdict: FLUX.2 [flex] is the clear winner as it provides a much more cinematic and detailed interpretation of the prompt, with lighting that perfectly matches a New York night scene. While Z-Image Turbo followed the prompt instructions, it suffered from lower image quality and a less convincing environment. FLUX.2 [flex] also succeeded better in capturing the specific 'bored' expression of the passenger.
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]
- + Excellent typography with perfect spelling in all fields
- + Highly cohesive vintage gothic aesthetic with subtle textures
- + The layout is balanced and easy to read as a formal invitation
- − The thorns in the border are less prominent than requested
- − Lighting is a bit flat compared to the requested cinematic look
Z-Image Turbo
- + Energetic composition with strong use of the thorns and webs in the border
- + Cinematic lighting on the pumpkin and background trees
- + Good use of parchment texture and layering
- − Spelling error in the location field showing 'The Archves'
- − The small scroll banner is split and the text is not placed on it as requested
- − Text layout feels a bit cramped and less elegant
Verdict: FLUX.2 [flex] wins because it successfully rendered all requested text accurately and integrated the scroll banner properly into the design. While Z-Image Turbo captured the 'cinematic' and 'thorns' aspects of the prompt more aggressively, its spelling error and poor scroll placement make it less functional as an invitation.
Bald man challenge
Image Editing“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”
AI Judge Analysis
FLUX.2 [flex]
- + Successfully added a full head of hair that follows the skull shape naturally.
- + Preserved the glasses, clothing, and background elements perfectly.
- + Maintained the original facial features and expression without distortion.
Z-Image Turbo
- + Maintained the general lighting and atmosphere of the scene.
- − Failed to add a full, thick head of hair as requested, only adding a very short buzz cut.
- − Removed the person's glasses, violating the requirement to preserve facial features.
- − Changed the background scenery significantly, replacing the desert shrubs with tall grass.
Verdict: FLUX.2 [flex] perfectly executed the edit by adding realistic hair while keeping every other detail of the source image intact. In contrast, Z-Image Turbo failed most of the prompt instructions: it did not provide a thick head of hair, it removed the glasses, and it unnecessarily altered the background. FLUX.2 [flex] is the clear winner for its superior edit accuracy and source preservation.
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]
- + Text is perfectly rendered and centered as requested.
- + Correctly identifies and displays the Japanese flag icon.
- + Excellent miniature 3D aesthetic with a variety of sushi types (nigiri and maki).
- − The diorama base is a bit plain, matching the background color exactly.
Z-Image Turbo
- + The 3D textures on the salmon and rice are very tactile and 'squishy' in appearance.
- + Good use of a multi-toned diorama base for better visual separation.
- − Incorrectly uses the Chinese flag instead of the Japanese flag.
- − The text 'SUSHI' is not centered under 'JAPAN' as requested.
- − The sushi composition is very basic with only one piece of nigiri.
Verdict: FLUX.2 [flex] is the clear winner as it followed all instructions, including the correct flag and text alignment. Z-Image Turbo failed a critical cultural context check by placing a Chinese flag next to the text 'JAPAN SUSHI', and it also struggled with text centering and layout balance.
Over-the-top cartoon caricature
Editing“Create a caricature of me and my job. Make it exaggerated and humorous, incorporating my profession as a tv show anchor and my love for dogs and hockey.”
AI Judge Analysis
FLUX.2 [flex]
- + Excellent adherence to the 'caricature' style with an exaggerated, humorous design.
- + Successfully incorporates all requested elements: TV anchor desk, multiple dogs, and hockey gear/jerseys.
- + Retains recognizable features from the source image subject while transforming the style.
- − The text on the news desk ('THEIANAES NEWS') is slightly garbled but legible.
Z-Image Turbo
- + Preserved the original photo quality and environment perfectly.
- − Completely failed to apply the caricature style or the requested job setting.
- − Missed the hockey theme entirely.
- − The only visible change is a very small, blurry dog in the background.
Verdict: FLUX.2 [flex] followed every instruction explicitly, creating a vibrant caricature that balanced the subject's likeness with the requested profession and hobbies. In contrast, Z-Image Turbo largely ignored the edit instructions, failing to change the style to a caricature or include any of the primary requested themes like hockey or the TV anchor setting.
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 adherence to the lighting requested, with atmospheric god rays and visible dew drops on the flowers.
- + Superior fur texture and detail across all four distinct animals.
- + Dynamic and balanced composition that captures the 'chasing' aspect of the prompt effectively.
- − The kitten is slightly less realistic compared to the other three animals.
- − The butterflies look a bit like stickers placed on top of the image.
Z-Image Turbo
- + Warm, pleasant color palette that captures the 'wholesome' vibe well.
- + Accurately includes all four requested animals in a tight, cute grouping.
- − Anatomical issues, particularly the golden retriever's paw merging into the rabbit's back.
- − Lower overall resolution and softer details compared to the masterpiece quality requested.
- − The 'god rays' are much less defined and the dew sparkles are represented as generic white dots.
Verdict: FLUX.2 [flex] is the clear winner as it successfully rendered all four distinct animals with high fidelity and captured the atmospheric elements like god rays and dew drops much more effectively. Z-Image Turbo struggled with animal anatomy and produced a softer, less detailed image with less dynamic composition.
Studio Ghibli Anime Style
Editing“Transform this photo into a Studio Ghibli–inspired illustration. Use soft pastel colors, hand-painted textures, gentle lighting, dreamy backgrounds, and a warm, nostalgic mood”
AI Judge Analysis
FLUX.2 [flex]
- + Perfectly captures the Studio Ghibli art style with cel-shaded characters and painterly backgrounds.
- + High edit accuracy while maintaining the composition and core identity of the meme.
- + Uses the requested soft pastel color palette and warm lighting effectively.
- − Changed the expression of the girlfriend from angry/indignant to happy, losing the original context of the meme.
Z-Image Turbo
- + Preserved the facial expressions and emotions from the original photo more accurately.
- − Failed to apply the requested artistic style, remaining almost entirely photographic.
- − No evidence of hand-painted textures or dreamy Ghibli-inspired backgrounds.
- − The lighting and colors remain realistic rather than pastel or nostalgic.
Verdict: FLUX.2 [flex] followed the creative instructions perfectly, transforming the photo into a beautiful Ghibli-style illustration with hand-painted textures and soft colors. In contrast, Z-Image Turbo largely ignored the edit instructions, producing a result that looks like a slightly filtered photograph rather than an illustration.
Golden Hour Stroll
Image Editing“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”
AI Judge Analysis
FLUX.2 [flex]
- + Excellent execution of blowing hair that looks energetic and natural.
- + Succesfully added many flying green leaves throughout the foreground.
- + Maintained high image fidelity and facial resemblance to the source.
- − The leaf placement looks slightly like a digital overlay rather than being part of the environment.
- − Minor anatomical distortion where the hand meets the dog's head.
Z-Image Turbo
- + Natural integration of autumn leaves into the existing scene.
- + Preserved the general composition and lighting of the original photo well.
- − Failed to significantly alter the hair to show 'blowing in the wind'.
- − The model significantly changed the subject's face, losing the likeness of the source image.
- − The overall sense of motion is much lower than requested.
Verdict: FLUX.2 [flex] is the clear winner as it successfully implemented both requests: dynamic blowing hair and flying leaves, while maintaining the subject's identity. Z-Image Turbo added some leaves but failed to animate the hair and completely altered the woman's face, failing the source preservation aspect of the edit.
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 text rendering for both the name and the banner
- + Elegant minimalist vector style with clean lines
- + Excellent composition with the curved text framing the cloche
- − The steam is a bit faint compared to the rest of the logo
Z-Image Turbo
- + Stronger vector presence with bold colors
- + Good use of the brown and cream color palette
- − The 'f f' in 'Caffè' are merged awkwardly
- − The steam icon looks unbalanced and disconnected
- − The horizontal banner is less dynamic than the curved ribbon in Model A
Verdict: FLUX.2 [flex] produced a superior logo with perfect typography and a more sophisticated composition that feels historically appropriate for the brand. Z-Image Turbo struggled with the letter spacing in the main title and created a less balanced graphic overall.
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 perfect spelling and clear hierarchy.
- + Followed the logical progression of all six requested steps in a structured layout.
- + Accurate NASA-inspired color palette and high-quality flat vector iconography.
- − The layout is a bit cramped at the bottom.
- − Missed the final 'Landing' step, stopping at 'Descent'.
Z-Image Turbo
- + Clean flat vector style and bold, readable icons.
- + Uses a simple, modern layout that is easy to scan.
- − Major spelling errors including 'APOLIO E 11', 'Translurian', and 'Descenty'.
- − Failed to include the specific step-by-step sequence requested, combining elements randomly.
- − The rocket design is generic and does not resemble a Saturn V.
Verdict: FLUX.2 [flex] successfully captures the sophisticated look of a professional infographic with accurate text and a logical flow, despite missing the final landing step. Z-Image Turbo suffers from significant spelling errors and fails to map out the specific mission steps requested in the prompt, resulting in a disconnected collection of icons. FLUX.2 [flex] is much more aligned with the 'NASA-inspired' professional aesthetic requested.
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