Distilled version of Black Forest Labs' FLUX.2 [dev] outperforming it at a cheaper price. Developed by fal.ai.
Settled by community votes across 10 shared challenges, with an AI judge weighing in on each.
FLUX.2 [dev] Turbo
#4 of 44 in Text-to-Image
Not enough comparable category data
The chart appears once both models have ratings across at least three shared arena categories.
Z-Image Turbo
#15 of 44 in Text-to-Image
Where the votes landed
FLUX.2 [dev] Turbo
100.0%
win rate
Ties
0.0%
Z-Image Turbo
0.0%
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.2 [dev] Turbo
- + Excellent adherence to the 'partially visible through the glass' instruction for the plant.
- + Highly realistic textures on the wood grain, glass imperfections, and book cover.
- + Very accurate lighting and reflections, including the blue sphere's reflection on the bottom glass pane.
- − The plant's leaves are quite sharp through the glass, slightly reducing the refractive distortion expected from thick glass.
Z-Image Turbo
- + Follows all basic prompt requirements regarding object placement.
- + Clean, minimalist composition with soft, pleasing lighting.
- − The plant is behind the cube but not visible *through* the glass as requested; it is only visible above/around it.
- − Lower overall texture detail on the wooden table and the book's spine compared to the competitor.
- − The bottom of the cube appears to be a solid mirror rather than clear glass.
Verdict: FLUX.2 [dev] Turbo followed the prompt much more accurately, specifically the requirement for the plant to be visible through the glass cube. It also displayed superior photorealism in the textures of the wood and the red book. Z-Image Turbo produced a clean image, but failed the specific spatial relationship between the plant and the glass cube.
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 all prompt details including motion blur, reflections, and 'repairing' action.
- + High level of realistic skin texture and facial detail.
- + Convincing rain and wet pavement effects that feel cinematic.
Z-Image Turbo
- + Natural skin tones and realistic lighting.
- + Good shallow depth of field effect.
- − Fails to show the man 'repairing' the bike; he is simply holding the handlebars.
- − Missing requested motion blur from passing cars.
- − Anatomical issues with the feet and awkward perspective on the background car.
Verdict: FLUX.2 [dev] Turbo followed the prompt much more accurately, capturing the specific 'repairing' action with tools on the ground and the requested motion blur of passing cars. Z-Image Turbo missed several key descriptors, such as the motion blur and the core activity of repairing the bicycle, resulting in a more static and less narrative image.
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
- + Exceptional detail on the engraved plate armor and leather straps.
- + Perfect adherence to specifically requested details like small beads in the braids.
- + Strong portrait composition with lifelike facial features and realistic skin textures.
- − The torch flame in the background looks slightly flat compared to the figure.
- − Bokeh sparks are a bit uniform in size across the frame.
Z-Image Turbo
- + Excellent light interaction between the torch and the armor's surface.
- + Atmospheric lighting and very natural-looking bokeh sparks.
- + Solid armor engraving and fabric textures.
- − The character appears slightly cross-eyed or has an inconsistent gaze.
- − The beads in the hair are much less prominent and detailed than requested.
- − Lower resolution/clarity in fine textures compared to the competitor.
Verdict: FLUX.2 [dev] Turbo provides a superior close-up portrait with incredible sharpness and perfect adherence to the prompt, particularly the small beads and intricate engravings. While Z-Image Turbo handles the play of torchlight and sparks with more artistic flair, it suffers from a slight eye alignment issue and lacks the crisp textural detail found in FLUX.2.
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 layout that balances high-quality imagery with clear text sections.
- + Strong typography rendering with distinct headers and legible body text.
- + Professionally structured grid that feels like a real commercial design.
- − The 'Mains' category contains mostly pizza images, showing slight logical inconsistency with the food shown.
- − Small text elements like prices and descriptions contain minor gibberish characters.
Z-Image Turbo
- + Strong minimalist aesthetic with bold, vibrant orange accents.
- + Good variety of dish types including pasta and salads, matching the Mediterranean menu theme.
- + Clean grid layout for the photography.
- − Major spelling error in the primary header ('PIZZA MANS').
- − The layout is slightly disjointed with text sections placed oddly beside and below images.
- − Several text artifacts and nonsensical headers like 'SE IIIION'.
Verdict: FLUX.2 [dev] Turbo produces a much more professional and realistic menu layout that closely follows the prompt's request for a clean, professional design. While Z-Image Turbo has a bold aesthetic, the significant spelling errors and awkward text placement make it less functional than FLUX.2, which feels like a production-ready design template.
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
- + Excellent authentic chalk texture with realistic smudges and dust
- + Perfect spelling for all requested menu items
- + Natural, consistent handwriting style that matches the prompt's requested 'elegant cursive'
- − The pricing for the Risotto is slightly disjointed with the dollar sign separated from the number
Z-Image Turbo
- + Legible and clean layout
- + Accurate interpretation of the date and prices
- − Typo in 'Mushroom' (rendered as 'Mustroom')
- − The chalk texture looks too digital/uniform compared to actual handwriting
- − Minimal blackboard smudging/realism in the background
Verdict: FLUX.2 [dev] Turbo significantly outperforms Z-Image Turbo in realism, capturing the true grainy texture of chalk and natural smudging on a used blackboard. Furthermore, FLUX.2 had perfect spelling across the entire board, whereas Z-Image Turbo introduced a typo in the first menu item.
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 prompt adherence with the capybara wearing the hat and jacket correctly.
- + The composition perfectly captures the 'scene inside a taxi' feel from the front windshield.
- + High detail on the capybara's paws and the businesswoman's bored expression.
- − The businesswoman is placed in the front passenger seat instead of the back seat as requested.
Z-Image Turbo
- + Successfully places the businesswoman in the back seat as requested.
- + Good lighting on the capybara's fur and the taxi uniform cap.
- − The capybara's paws are not actually touching the steering wheel, appearing to float behind it.
- − The steering wheel is positioned strangely in the middle of the dash rather than in front of the driver.
Verdict: FLUX.2 [dev] Turbo produces a much more convincing and high-quality image with better anatomical placement, though it failed the 'back seat' instruction by placing the passenger in the front. Z-Image Turbo followed the spatial instruction for the passenger's seat but failed significantly on the logic of the car's interior, specifically the steering wheel and the driver's grip.
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 for all requested details, including date and location.
- + Superior composition with a cohesive illustrative style that feels like a vintage poster.
- + Integrated the scroll banner effectively into the bottom layout.
- − The parchment effect is restricted to the corners rather than being the primary background for the text.
Z-Image Turbo
- + Strong parchment texture throughout the design reflecting the vintage request.
- + Good use of shadows and highlights on the Jack-o-lantern for depth.
- + Bold, atmospheric gothic font choices.
- − Typos in the text, specifically 'Archves' instead of 'Arches'.
- − The scroll banner for the secondary text was not rendered correctly, appearing as floating scrolls in the middle.
- − The placement of the 'You are invited...' text is tiny and cramped at the top.
Verdict: FLUX.2 [dev] Turbo provided a significantly better result by rendering all text perfectly without typos and creating a balanced, professional composition. Z-Image Turbo struggled with the specific text requirements, including a spelling error in the location and a disjointed layout of the decorative scroll elements.
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, especially the salmon and rice grains.
- + Correctly depicted the Japanese flag icon as requested.
- + Balanced composition with tasteful garnishes like wasabi and ginger.
- − The text 'JAPAN' is slightly off-center compared to the 'SUSHI' text below it.
Z-Image Turbo
- + Clean, soft-styled 3D cartoon aesthetic that fits the 'miniature' prompt.
- + Good adherence to the isometric perspective and raised diorama base.
- − Incorrectly used the flag of China instead of the Japanese flag.
- − The texture of the salmon is very plastic-looking compared to the requested realistic PBR materials.
- − The rice structure appears somewhat amorphous and less defined.
Verdict: FLUX.2 [dev] Turbo followed all prompt instructions, including the specific flag icon, and provided a much more sophisticated level of detail in the textures. Z-Image Turbo failed a key instruction by generating the Chinese flag for a Japanese-themed scene and had lower-quality material rendering. FLUX.2 [dev] Turbo is the clear winner for its superior visual quality and prompt adherence.
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
- + Excellent adherence to all prompt elements, including the specific golden sunrise and god rays.
- + Superior fur texture and detail that truly feels '8K masterpiece' quality.
- + Dynamic and natural composition with playful interactions between the animals.
- − The kitten's tail looks slightly detached or oddly positioned in the grass.
- − Some of the floating water droplets are a bit excessive.
Z-Image Turbo
- + Captures all four requested animals correctly.
- + Bright and cheerful lighting that fits the wholesome vibe.
- + Cute character design for the animals.
- − Lower overall resolution and softer textures compared to Model A.
- − The butterflies look flat and lack the photorealistic detail of the rest of the scene.
- − Anatomy issues where the puppy's paw is unnaturally clipping into/onto the bunny's back.
Verdict: FLUX.2 [dev] Turbo significantly outperforms Z-Image Turbo in terms of visual fidelity, lighting, and realistic textures. While both models captured all four requested animals, FLUX.2 delivered a much more professional, 8K-style render with beautiful god rays and intricate fur details, whereas Z-Image had several anatomical clipping issues and a softer, less detailed finish.
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
- + Excellent adherence to the 'vintage' and 'cloche' aesthetic with high-quality textures.
- + Perfect typography including the accent on 'Caffè' and a well-designed banner.
- + Superior composition with an authentic vector emblem feel and subtle paper texture.
- − The steam lines are a bit thick compared to the rest of the fine detail.
Z-Image Turbo
- + Clean, minimalist layout that meets the basic requirements of the prompt.
- + Successfully includes all text elements and the cloche icon.
- − The 'Est. 1720' is placed in a flat bar rather than an elegant banner as requested.
- − The steam and cloche details look a bit generic and clip-art-like compared to Model A.
- − The typography for 'Caffè Florian' is less integrated into the emblem style.
Verdict: FLUX.2 [dev] Turbo provided a much more professional and authentic vintage logo design, featuring a beautifully rendered banner and textured background that matched the minimalist vector style perfectly. Z-Image Turbo followed the instructions but produced a flatter, more generic graphic that lacked the artistic polish and sophisticated typography found in the first image.
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