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
#47 of 62 in Text-to-Image
LongCat-Image
#62 of 62 in Text-to-Image
Where the votes landed
FLUX.1 [schnell] FP8
100.0%
win rate
Ties
0.0%
LongCat-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 photorealism and lighting
- + Clean architectural glass design
- + Vibrant colors and sharp focus
- − The glass object is a rectangular prism rather than a cube
- − The blue sphere is floating on a shelf rather than resting on the base of the cube
LongCat-Image
- + Perfect adherence to the 'cube' shape
- + Accurately places the sphere inside on the floor of the cube
- + Captures the soft window lighting perfectly
- − The glass has a slight cyan tint that wasn't requested
- − The red book edges are slightly less crisp than Model A
Verdict: While FLUX.1 [schnell] FP8 produces a more striking and high-resolution commercial-style photograph, it fails to generate a cube, instead creating a tall prism. LongCat-Image follows all spatial instructions perfectly, providing a true glass cube with the sphere in the correct position.
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 handling of wet pavement reflections and light quality
- + Superior skin texture and facial realism
- + Strong cinematic color grading and composition
- − Fails to depict light rain visible in the air
- − Lack of motion blur on passing cars as requested
LongCat-Image
- + Clearly depicts falling rain and its interaction with surfaces
- + Captures the 'repairing' action more accurately through the crouching pose
- + Good integration of background elements like Japanese signage
- − Anatomical and structural issues with the bicycle wheels and frame
- − Faces and clothing textures are slightly waxy compared to a natural photo
- − Missed the motion blur requirement for passing cars
Verdict: FLUX.1 [schnell] FP8 produces a much more convincing and high-quality image of a person, looking like a real 50mm photograph, though it fails to render the actual falling rain or motion blur. LongCat-Image follows the situational prompt better by showing the rain and a 'repairing' posture, but it suffers from significant AI artifacts in the bicycle's geometry and less realistic skin textures. FLUX.1 [schnell] FP8 is the winner for its superior visual fidelity and photographic realism.
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
- + Extremely high skin texture detail and lifelike eyes
- + Excellent dramatic lighting following the torchlight prompt
- + Highly effective shallow depth of field
- − Armor is largely cropped out and less detailed than the face
- − Braids are minimal and hard to distinguish
- − Contains some nonsensical text artifacts in the bottom right corner
LongCat-Image
- + Excellent adherence to all prompt elements including braids and beads
- + Beautifully engraved plate armor with clear leather and cloth textures
- + Strong composition showing the full upper torso and battle-worn features
- − Loss of detail in the face compared to the skin texture of Model A
- − The scars appear slightly artificial or painterly
- − The depth of field is less shallow than requested
Verdict: LongCat-Image captured the full essence of the paladin prompt, including specific details like the beads in the braids and the engraved armor. While FLUX.1 [schnell] FP8 produced a more realistic skin texture and dramatic lighting, it failed to show the armor effectively and included stray text artifacts.
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
- + Excellent clean layout that resembles a real world menu.
- + Clear sections for Appetizers, Pizza, and Mains as requested.
- + The food photos are well-integrated into a legible grid.
- − Several spelling errors in headings like 'Apptizers', 'Orceters', and 'Seccer'.
- − The text for individual items is just placeholder gibberish.
LongCat-Image
- + High-quality food photography with good color and detail.
- + Strong use of vibrant accents and bold fonts.
- + Creative use of color blocking for a modern feel.
- − The layout is cluttered and difficult to read as a functional menu.
- − Text rendering is poor with significant letter distortion.
- − Fails the 'white background' requirement by using large blocks of yellow and blue.
Verdict: FLUX.1 [schnell] FP8 is the clear winner for its superior composition and professional layout which perfectly matches the 'minimalist' and 'grid' requirements of the prompt. While LongCat-Image has high-quality photos, the layout is messy and ignores the specific instruction for a white background, resulting in a design that is less functional as a menu.
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
- + Features a very high-quality photorealistic rendering of the burger ingredients.
- + Includes the secondary message 'LIMITED TIME ONLY' as requested.
- − Significant text errors including spelling repetitions ('LIIMITED', 'NEEY') and price inaccuracies.
- − The starburst element is a dull grey and lacks the requested fiery glowing effect.
LongCat-Image
- + Excellent adherence to stylistic cues with vibrant, fiery glowing text as requested.
- + Perfect text rendering for all requested elements, including the price in a glowing starburst.
- + Better cinematic composition with the burning embers/charcoal at the base.
- − The burger is less 'exploded' than Model A's interpretation, appearing mostly intact.
- − The sauce texture looks a bit plastic or artificial compared to the other ingredients.
Verdict: While FLUX.1 [schnell] FP8 offers a slightly more realistic food texture, it suffers from significant spelling errors and fails to apply the requested fiery style to the text. LongCat-Image followed the prompt instructions much more accurately, delivering perfect typography, the correct price, and the specific glowing aesthetic requested for the advertisement.
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 text rendering with very few spelling errors
- + Accurately follows the specific list of menu items and dates
- + The chalk texture and handwriting style appear authentic and natural
- − Layout is slightly cluttered with some repeated lines of text
- − The handwriting is neat but lacks the 'elegant cursive' style requested for the title
LongCat-Image
- + Strong aesthetic appeal with a professional café-style layout
- + The title font is more stylized and artistic as requested
- + Excellent chalk dust effects on the board edges
- − Significant spelling errors and gibberish text throughout the menu items
- − Fails to accurately render the specific menu item names from the prompt
- − Text legibility is poor compared to Model A
Verdict: FLUX.1 [schnell] FP8 is the clear winner due to its superior ability to render the specific text requested in the prompt with high accuracy and legibility. While LongCat-Image creates a more visually pleasing café environment and artistic title, the actual menu content is mostly unintelligible gibberish, failing the core requirement of the text-to-image challenge.
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
- + Successfully followed the difficult 'horse on top' prompt instruction
- + Cinematic lighting and high detail consistency
- + Surreal and creative composition
- − Anatomy and equipment are slightly abstract/messy
- − The 'astronaut' is represented more as a piece of equipment than a person in a suit
LongCat-Image
- + Technically clear rendering of an astronaut and a horse
- + Good space background scenery
- − Failed the primary prompt instruction of having the horse on top
- − Lower visual quality with artifacts in the sky and floating objects
- − Generic interpretation of a common concept
Verdict: FLUX.1 [schnell] FP8 correctly interpreted the specific and unusual instruction of having the horse ride the astronaut, resulting in a truly surreal image. LongCat-Image ignored the spatial instruction entirely, producing a standard astronaut-riding-horse image with several digital artifacts in the background. FLUX is the clear winner for adherence and creativity.
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
- + Effective lighting and cinematic depth-of-field.
- + Capybara expression is calm and professional as requested.
- + The lady's expression captures the requested 'bored' vibe perfectly.
- − The lady strangely appears to be holding two phones.
- − The capybara's hands are a bit muddled around the steering wheel.
- − The taxi cap looks more like a plastic toy than a uniform hat.
LongCat-Image
- + Excellent anatomical detail on the capybara's paws and fur.
- + Realistic taxi driver cap and jacket sleeve details.
- + Sharp, high-resolution texture across the entire image.
- − The capybara's head is excessively large, breaking the sense of scale.
- − There are two women in the back instead of one as requested.
- − The women's faces look slightly distorted and AI-generated compared to the animal.
Verdict: FLUX.1 [schnell] FP8 creates a more atmospheric and humorously accurate composition, perfectly nailing the bored expression of the passenger, despite a weird glitch involving two phones. LongCat-Image has superior technical texture and capybara anatomy, but fails the prompt instructions by including two passengers and having significant scaling issues with the capybara's head size.
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
- + Atmospheric cinematic lighting and nice gothic font style for the title.
- + Excellent spooky background composition with silhouettes of twisted trees.
- − Poor text rendering with numerous spelling errors and garbled words in the body text.
- − Missing the thorns and webs explicitly mentioned for the border.
LongCat-Image
- + Excellent adherence to the border requirements, including webs and thorns.
- + High-quality text rendering for the scroll and title with minimal errors.
- + Central illustration of the jack-o-lantern and background is very crisp.
- − The parchment cutout has a slightly sharp, digital edge that looks less 'vintage'.
- − Minor spelling error in the location name ('The Armiees' instead of 'The Arches').
Verdict: LongCat-Image is the clear winner as it successfully incorporated all the complex prompt elements, including the thorny border and specific text phrases, whereas FLUX.1 [schnell] FP8 struggled significantly with the body text legibility. While FLUX.1 had a moodier atmosphere, LongCat-Image provided a much more functional and accurate invitation layout.
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 isometric diorama perspective
- + Clean PBR-style textures and soft lighting
- + Good interpretation of the 'miniature' aesthetic
- − Text is incorrect and fails to include the word 'SUSHI'
- − The flag icon is integrated poorly into the text
- − Rice texture looks slightly unnatural compared to the toppings
LongCat-Image
- + Perfect text rendering of 'JAPAN' and 'SUSHI'
- + Includes a clear flag icon as requested
- + High-quality texture on the fish and a more realistic rice grain representation
- − The perspective is a bit lower than a true 45-degree isometric top-down view
- − Background has slight noise/texture instead of being a solid flat color
Verdict: While FLUX.1 [schnell] FP8 captured the isometric diorama style more accurately, it failed significantly on the text requirements, producing garbled words. LongCat-Image followed the prompt instructions perfectly, including the specific text and flag, while maintaining a very high visual quality for the sushi models. LongCat-Image is the clear winner for its superior 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.1 [schnell] FP8
- + Excellent fur texture and lighting integration
- + Expressive facial features and coherent artistic style
- + Composition feels naturally crowded and joyful as requested
- − Failed to include a rabbit, instead generating multiple kittens
- − Anatomical oddities on the smallest animal creature in the bottom right
- − Fox kit looks more like a ginger cat/fox hybrid
LongCat-Image
- + Successfully included the golden retriever and fox kit with distinct features
- + Very clear 'god rays' lighting effect and dew sparkles
- + Sharp, high-resolution details on the fur and flowers
- − The cat and rabbit were merged into a single bizarre hybrid animal with rabbit ears on a kitten body
- − Misses one of the four required distinct animal types by merging two together
- − Composition feels a bit more staged and less 'tumbling' than the prompt suggested
Verdict: Both models failed to correctly provide four distinct species of animals. FLUX.1 [schnell] FP8 produced a more aesthetically pleasing and 'masterpiece' quality image with superior lighting and fur texture, despite missing the rabbit entirely. LongCat-Image attempted all animals but created a confusing kitten-rabbit hybrid, which significantly detracts from the realism of the scene.
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
- + Clean vector-style execution
- + Balanced composition with good symmetry
- + Accurate date placement on the banner
- − Failed significantly on the text spelling (FLAMILAN instead of Florian)
- − Interpreted the cloche as an architectural dome
LongCat-Image
- + Correctly spelled 'Caffè Florian'
- + Excellent interpretation of the cloche dome with steam
- + Strong vintage texture and aesthetic
- − Repetitive text ('Caffè' appears twice)
- − Composition feels a bit crowded and heavy compared to Model A
Verdict: LongCat-Image is the clear winner because it correctly spelled the requested text and accurately depicted a cloche dome, whereas FLUX.1 [schnell] FP8 failed the spelling and visual concept. While FLUX.1 provided a cleaner vector look, LongCat-Image's superior adherence to the prompt and better use of texture makes it the more effective 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 typography rendering for the main title.
- + Clean and modern layout that follows the 'infographic' request effectively.
- + Adheres well to the requested navy, white, and muted red palette.
- − Icons are overly abstract and don't clearly represent the specific mission phases.
- − Text below headings is mostly gibberish.
- − The timeline flow is a bit repetitive and loses clarity toward the end.
LongCat-Image
- + Illustrative style is engaging and clearly depicts space-related imagery.
- + Icons for Earth and the Lunar Module are more recognizable than in the competing image.
- + Good use of the color palette.
- − Failed to render the 'APOLLO 11' text correctly, resulting in 'Aaqo 11'.
- − Composition feels cramped and follows a vertical split rather than a clear sequential infographic flow.
- − Inaccurate iconography, such as using Space Shuttle-style boosters instead of the Saturn V.
Verdict: FLUX.1 [schnell] FP8 produced a superior infographic layout with clean lines and professional typography, though its iconography was a bit generic. LongCat-Image provided more detailed illustrations but failed on text accuracy and followed a less logical information hierarchy for a multi-step poster.
Explore each model
6B parameter image generation model excelling at rendering multilingual text directly in generated images