FP8 quantized variant of Black Forest Labs' FLUX.1 [schnell] model, offering ~2x faster inference with reduced precision while maintaining high-quality image generation in 4 steps
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX.1 [schnell] FP8
#46 of 62 in Text-to-Image
Qwen Image
#40 of 62 in Text-to-Image
Where the votes landed
FLUX.1 [schnell] FP8
100.0%
win rate
Ties
0.0%
Qwen Image
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.1 [schnell] FP8
- + Excellent refraction and caustic effects through the glass
- + High resolution and sharp details on the sphere and table texture
- + Accurate soft lighting and depth of field
- − The glass container is more of a tall rectangular prism than a 'cube'
- − The sphere is levitating on an internal shelf rather than resting on the bottom
Qwen Image
- + Perfect adherence to the 'cube' shape
- + Realistic placement of the sphere at the bottom of the cube
- + Clean, balanced composition
- − Lighting is a bit flat compared to Model A
- − The plant lacks the refraction/visibility through the glass requested in the prompt
- − Minor structural artifacts in the internal glass reflections
Verdict: Both models followed the prompt accurately, but FLUX.1 [schnell] FP8 produced a more visually stunning image with superior glass physics and lighting. However, Qwen Image followed the specific geometric instruction for a 'cube' better than FLUX, which rendered a rectangular prism.
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 [schnell] FP8
- + Excellent skin texture and facial details
- + Strong reflection rendering on the wet asphalt
- + Vivid colors and high cinematic quality
- − Missed the 'motion blur' request for the background vehicles
- − The man appears to be posing with/holding the bike rather than actively repairing it
- − The background sign language and car logos are slightly garbled
Qwen Image
- + Better adherence to the requested motion blur of passing cars
- + More accurate representation of light rain with visible falling droplets
- + Authentic Japanese street atmosphere
- − The bike has anatomical/mechanical errors, specifically the pedal and kickstand logic
- − Skin textures are slightly smoother and less detailed than model A
- − The subject's hands are somewhat blurry and lack definition
Verdict: FLUX.1 [schnell] FP8 wins on raw image quality and character detail, specifically the natural skin texture and the richness of the lighting. However, Qwen Image does a significantly better job at capturing the requested 'motion blur' and 'rain' effects, creating a more dynamic scene despite the mechanical errors in the bicycle's design.
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 [schnell] FP8
- + Exceptional skin texture and lifelike eye detail
- + Excellent implementation of torchlight reflection on the face and armor
- + Superior close-up composition with high cinematic appeal
- − Missed the request for 'hair braided with small beads'
- − Armor and leather details are partially obscured by the tight framing
Qwen Image
- + Strong adherence to all prompt elements including braided hair with beads and leather straps
- + Clearly visible ornate engraved plate armor
- + Good inclusion of bokeh sparks and a physical torch
- − Skin and hair texture look slightly artificial and plastic-like
- − Lighting is less integrated with the subject compared to Model A
- − Scars look like surface paint rather than realistic wounds
Verdict: FLUX.1 [schnell] FP8 produces a significantly more lifelike and high-quality image with impressive lighting and skin textures, though it fails to include the requested hair braids and beads. Qwen Image adheres much more closely to the specific details of the prompt like the armor and hair accessories, but the overall visual quality and realism are notably lower. Flux is preferred for its artistic and technical quality, while Qwen is better for literal prompt adherence.
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 [schnell] FP8
- + Cluttered but realistic layout that looks like a functional two-page menu.
- + Includes specific sections for pizza and mains as requested.
- + Clean white background with professional alignment of text and prices.
- − Text contains frequent misspellings like 'ORCETERS' and 'APPTIZERS'.
- − The food photos lack the 'vibrant' pop requested, looking somewhat muted and repetitive.
- − The 'grid' requested is applied to the whole page rather than being a distinct design element for photos.
Qwen Image
- + Strong adherence to the 'grid' and 'vibrant accents' request with colorful backgrounds for photos.
- + Excellent bold sans-serif typography that feels truly modern and minimalist.
- + High-quality, appetizing food photography that fits the casual dining vibe.
- − Failed to provide a distinct section for 'mains', grouping them with pizza instead.
- − Text rendering becomes illegible in the smaller menu item descriptions.
- − One grid cell is just a solid orange block with garbled text rather than a photo.
Verdict: Qwen Image adheres better to the 'modern minimalist' aesthetic and provides the vibrant grid layout requested, resulting in a more visually appealing design. However, FLUX.1 [schnell] FP8 produced a more structurally complete menu with all requested sections, even though the typography and spelling were less polished.
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 [schnell] FP8
- + Naturalistic bun texture and lighting on the food
- + Good use of negative space for the central subject
- − Severely failed text rendering with multiple spelling errors and repetitions
- − Incorrect price was featured prominently (€69 vs €6.99)
- − Minimal 'fiery' effect on the text itself
Qwen Image
- + Excellent adherence to styling with fiery, glowing text effects
- + Perfect text accuracy including the currency and secondary message
- + Highly dynamic composition with various exploded ingredients
- − The burger structure is a bit more 'stacked' than 'exploded' compared to the pieces in the air
- − Slight overlap of the starburst and the burger bun
Verdict: Qwen Image is the clear winner as it followed every detail of the prompt, especially the difficult task of rendering fiery, glowing text and the correct price within a starburst. FLUX.1 [schnell] FP8 failed significantly on the text, producing misspelled words and the wrong price point.
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 [schnell] FP8
- + Excellent realistic chalk texture on the board.
- + Good adherence to the request for natural variations in letter size.
- − Numerous spelling errors including 'Risortto', 'Griffled', and 'Octemon'.
- − Text layout is cluttered with repetitive lines and messy price alignment.
Qwen Image
- + Near-perfect spelling of complex menu items.
- + Consistent and aesthetically pleasing handwriting style across the board.
- + Great background composition that establishes the cozy café setting.
- − Small error in the date, rendering '20026' instead of '2026'.
- − The text appears slightly too perfect/digital compared to the requested raw chalk texture.
Verdict: Qwen Image is the superior choice because it successfully renders the requested text with high accuracy and professional layout, despite a minor typo in the year. FLUX.1 [schnell] FP8 captures the requested chalk texture better but fails significantly on text legibility and spelling, producing a messy and nonsensical list of items.
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 [schnell] FP8
- + Attempts the specific role-reversal requested in the prompt
- + High level of texture detail on the horse hair and equipment
- + Dramatic cinematic lighting and composition
- − Anatomy becomes messy with two horse heads and fused bodies
- − The 'astronaut' is represented as equipment rather than a clear figure
Qwen Image
- + High clarity and clean rendering
- + Good anatomical representation of a horse
- + Consistent lighting
- − Completely ignores the negative constraint that the horse should be on top
- − Produces a generic interpretation of the prompt
Verdict: This was a difficult prompt involving a logic reversal. FLUX.1 [schnell] FP8 followed the specific instruction to place the horse on top, even though the resulting anatomy is surreal and somewhat incoherent. Qwen Image completely failed the core instruction of the prompt, producing a standard astronaut riding a horse, which disregarded the 'not vice versa' requirement.
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 [schnell] FP8
- + Excellent texture on the capybara's fur.
- + Consistent interior lighting with the yellow cab theme.
- + Successfully captures the 'bored' expression of the passenger.
- − The passenger is holding two phones simultaneously, which is a hallucination.
- − The taxi cap is a small plastic-looking piece rather than a traditional driver's hat.
- − The passenger is on the wrong side relative to the driver for a standard taxi.
Qwen Image
- + Features a more authentic taxi driver cap and professional jacket.
- + Proper placement of the passenger in the back seat with a realistic bored expression.
- + Higher realism in the vehicle's interior hardware and framing.
- − The capybara's hands/paws look slightly mutated and more like primate hands.
- − Text on the roof sign is garbled ('YOXI').
Verdict: Qwen Image is the winner because it follows the spatial composition and costuming requirements much more logically than FLUX.1 [schnell] FP8, which depicted the passenger holding two phones and sitting in the front passenger seat. Although Qwen Image has some issues with the anatomy of the capybara's paws, the overall scene coherence and professional attire of the driver better match the prompt.
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 [schnell] FP8
- + Features a vibrant, glowing jack-o-lantern.
- + Captures the 'dark parchment' texture well in the border.
- − Severely garbled text and typos throughout the invitation details.
- − Missed the specific requested border elements like spider webs.
Qwen Image
- + Excellent adherence to border details, including webs and thorns.
- + Perfectly rendered event details at the bottom with no typos.
- + Stronger gothic aesthetic with twisted trees and cinematic lighting.
- − Small typo in the main header text ('Halle Party' instead of 'Halloween Party').
Verdict: While FLUX.1 [schnell] captures the color of the pumpkin well, it fails significantly on text legibility and following the border instructions. Qwen Image provides a much more cohesive and high-quality design, correctly placing the event details and incorporating the complex border elements requested in the prompt, despite a minor typo in the largest header string.
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 [schnell] FP8
- + Excellent 3D modeling effect with soft, clean textures
- + Perfectly executed 45-degree isometric perspective
- + High clarity and centered composition on a professional diorama base
- − Failed the text prompt, spelling 'JAPAN' and 'JAPAN' (with a dot) instead of 'SUSHI'
- − The flag icon is integrated into the text incorrectly
Qwen Image
- + Perfect adherence to text instructions including 'JAPAN' and 'SUSHI'
- + Includes the flag icon as requested in both the text area and the scene
- + Accurate representation of the diorama and miniature cartoon style balance
- − Chopsticks are slightly melded into the rest of the plate
- − Background color is a bit saturated for 'solid light blue' compared to Model A
Verdict: While FLUX.1 [schnell] FP8 produced a more polished 3D render with superior materials, it failed significantly on the text rendering by repeating 'JAPAN' incorrectly. Qwen Image followed every instruction perfectly, including the text layout and the specific flag icon placement, while still maintaining high visual quality.
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 [schnell] FP8
- + Warm and vibrant lighting with effective god rays
- + High fur detail and texture
- + Great atmospheric depth with blurred foreground wildflowers
- − Failed to include a bunny, substituting it with extra cats/unidentified kits
- − The animal grouping looks congested and slightly unnatural
- − One cat has a noticeably distorted mouth expression
Qwen Image
- + Successfully included all four requested animals (dog, cat, bunny, fox)
- + Clearer 'chasing' action and dynamic posing
- + Better adherence to the 'dew sparkles' instruction with visible bokeh highlights
- − The fox's front paws look a bit like dark blobs or shadows
- − Lighting is a bit flatter compared to the intense golden glow of Model A
Verdict: Model B (Qwen Image) is the clear winner because it correctly followed the specific species requirements of the prompt, including the bunny which Model A omitted. While FLUX.1 [schnell] FP8 offered beautiful lighting and fur textures, its failure to generate all requested animals and the anatomical issues with the cat's mouth make it less successful overall.
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 [schnell] FP8
- + Strong vintage aesthetic with detailed scrollwork
- + Colors are well-balanced and authentic to the retro theme
- + Includes the 'Est. 1720' text clearly
- − Major spelling error in the main brand name ('Afe Flamilan')
- − The cloche looks more like a building dome than a food cloche
Qwen Image
- + Successfully rendered the cloche dome with steam
- + Minimalist vector style aligns well with the prompt
- + Text is closer to the correct spelling despite layout issues
- − Text layout is cluttered and difficult to read ('FloraiAN')
- − The typography lacks the 'classic' elegance requested in the prompt
Verdict: Both models struggled with the specific text 'Caffè Florian', but FLUX.1 [schnell] FP8 produced a more visually cohesive and professionally designed vintage emblem despite the significant spelling error. Qwen Image followed the literal prompt elements like the steam and the cloche dome more accurately, but the resulting typography and layout are messy and less aesthetically pleasing for a 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.1 [schnell] FP8
- + Excellent adherence to the 'flat-vector' style with crisp lines and balanced weight.
- + Sophisticated, professional layout that feels like a real educational infographic.
- + Strong adherence to the requested NASA-inspired color palette.
- − Text is mostly gibberish or nonsensical placeholder-style characters.
- − Confused the specific steps, only showing 5 main icons for 6 requested steps.
- − Iconography is a bit abstract and doesn't clearly represent the Saturn V rocket.
Qwen Image
- + High degree of legibility for step titles and names.
- + Correctly followed the numeric sequence of steps (though mislabeled some).
- + Individual icons like the Lunar Module and Saturn V are more recognizable as the objects described.
- − Poor composition with overlapping elements and a cluttered, non-linear flow.
- − Failed to follow the logical sequence of the mission (e.g., placing the Landing before the LEM on the surface).
- − Included the prompt text 'Stop at landing' as a header in the image itself.
Verdict: FLUX.1 [schnell] FP8 produced a much more aesthetically pleasing and professionally styled vector poster that adheres perfectly to the requested art style, even though the text is garbled. Qwen Image succeeded in making the text readable and the icons recognizable, but the overall layout is messy and the mission logic is flawed. FLUX.1 is the preferred choice for a designer who can easily replace the text.
Explore each model
Alibaba's Qwen image model