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
HiDream I1 Fast
#50 of 62 in Text-to-Image
Where the votes landed
FLUX.1 [schnell] FP8
0%
win rate
Ties
0%
HiDream I1 Fast
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 realization of light and depth
- + The 'partially visible through glass' requirement is handled with sophisticated refraction
- + The objects look very high-end and detailed
- − The glass enclosure is more of a rectangular prism than a perfect cube
- − The blue globe is resting on an internal shelf that wasn't requested
HiDream I1 Fast
- + Perfect adherence to the object arrangement and geometry
- + The blue sphere is solid and opaque as one might expect from the prompt
- + Accurate cube geometry
- − The red book lacks a white page edge, looking like a solid block
- − Reflections on the sphere are slightly inconsistent with the glass container
Verdict: Both models followed the complex spatial prompt very well. FLUX.1 [schnell] FP8 produced a more aesthetically pleasing, semi-transparent blue globe with complex refractions, though it added an extra internal shelf. HiDream I1 Fast followed the literal geometry of a 'cube' better and kept the sphere centered on the floor of the cube, making it the more accurate representation of the scene instructions.
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 lights
- + Superior skin texture and realism on the subject
- + Higher resolution and clearer details on the bicycle mechanics
- − Failed to include motion blur on passing cars as requested
- − Environmental colors feel slightly more generic than specifically Japanese
HiDream I1 Fast
- + Better captures the atmosphere of a Japanese street with misty background
- + Strong adherence to the wet pavement and lighting request
- + Good composition that feels more cinematic
- − Anatomical and structural issues with how the man is sitting/merging into the bike
- − Image lacks the sharpness and texture detail requested
- − The red bicycle frame has inconsistent geometry
Verdict: FLUX.1 [schnell] FP8 is the winner due to its significantly higher technical quality and realistic skin textures, despite missing the motion blur request. HiDream I1 Fast creates a nice mood and atmosphere, but suffers from major structural issues where the man's legs and the bicycle frame merge in an unrealistic way.
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 realism with visible pores and fine wrinkles
- + Intense and convincing lighting interact with the facial contours
- + Very sharp and lifelike eye details
- − Missed the prompt for braided hair with beads
- − The armor engraving is less distinct and slightly muddled
- − The framing is cropped very tightly, losing some of the 'paladin' context
HiDream I1 Fast
- + Excellent adherence to all prompt elements including braids, beads, and scars
- + Ornate engraving on the plate armor is highly detailed and clear
- + Great use of bokeh with visible torchlight and sparks in the background
- − Hair braids appear somewhat artificial and overly symmetrical
- − The skin texture has a slightly smooth, digital feel compared to Model A
- − The scars look like red paint rather than 'faint' healed tissue
Verdict: FLUX.1 [schnell] FP8 wins on raw photographic realism and facial detail, but it misses several key descriptive elements of the prompt like the braided hair and beads. HiDream I1 Fast provides much better prompt adherence and a superior composition for a paladin character, even if its textures are slightly more stylized and less lifelike than FLUX.1.
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 typography layout resembling a real menu
- + Clean and professional white background style
- + High number of clear, distinct food photos that follow the grid request
- − Nonsense words in section headers like 'ORCETERS' and 'SECCER'
- − The top 'MENU' text is partially cut off
- − Some food images look repetitive in composition
HiDream I1 Fast
- + Strong vibrant accents with the red header bar
- + Clearer pizza imagery with better crust details
- + Sans-serif font aligns well with the 'modern' request
- − Large garbled text at the top ruins the professional feel
- − Failed to include sections for 'Mains' or distinct sections as requested
- − Lower resolution and visible AI artifacts in the text characters
Verdict: FLUX.1 [schnell] FP8 is the clear winner as it successfully creates a multi-page menu layout with all requested sections (Appetizers, Pizza, Mains) and a professional grid. While HiDream I1 Fast has more vibrant colors, it failed the layout requirements and produced significant text corruption, particularly in the large header.
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
- + High resolution and realistic texture on the burger bun and patty.
- + Effective use of negative space with clear, bold primary text.
- − Major spelling errors in the secondary text lines.
- − The 'exploded' effect uses generic red and orange chunks rather than recognizable burger ingredients.
- − Significant price error with the giant yellow text.
HiDream I1 Fast
- + Excellent adherence to the 'fiery, glowing effect' for all text elements.
- + Includes recognizable burger components like cucumbers/pickles and onions in the suspension.
- + Accurate price rendering in the required starburst shape.
- − Minor spelling glitch in the circular secondary text element.
- − The burger composition looks a bit vertically compressed rather than truly exploded.
Verdict: HiDream I1 Fast followed the prompt's stylistic instructions much more closely, specifically the fiery text effects and the correctly priced starburst. While FLUX.1 [schnell] FP8 had higher base image quality, its text was riddled with spelling errors and it failed to render the price correctly.
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
- + The chalk texture on the individual letters is highly realistic and convincing.
- + Correctly followed the date requirement 'April 30, 2026'.
- + The wooden frame and lighting on the board look professionally photographed.
- − The text has significant spelling errors and repetitive lines which clutter the board.
- − Failed to provide cursive for the title as requested.
- − The layout becomes chaotic toward the bottom of the board.
HiDream I1 Fast
- + Successfully rendered the title in elegant cursive handwriting.
- + Included a much more 'cozy café' atmosphere with a background and foreground props.
- + The layout is clean and resembles a purposeful menu design.
- − Several text overlaps where words and numbers are superimposed on each other.
- − Slight spelling errors in menu items like 'Risott' and 'Buter'.
- − Chalk texture looks a bit more like a digital brush than physical chalk compared to Model A.
Verdict: HiDream I1 Fast is the preferred model because it captured the 'cozy café' ambiance and correctly applied cursive to the title, despite some overlapping text artifacts. FLUX.1 [schnell] FP8 produced superior chalk texture and better spelling for the date, but failed on the cursive requirement and produced nonsensical, repetitive text for the menu 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
- + Excellent adherence to the 'horse on top' spatial instruction.
- + Stunning cinematic lighting and high-quality textures on the Earth and horses.
- + Captures the surreal nature of the prompt very effectively.
- − The astronaut is represented as mechanical equipment rather than a clear humanoid figure.
- − Anatomic glitch with two horse heads/bodies merged into one unit.
HiDream I1 Fast
- + Natural horse anatomy and clear astronaut rendering.
- + Good lighting on the desert terrain.
- − Total failure to follow the 'horse on top' instruction.
- − Ignored the 'in space' setting, placing the subject in a desert.
- − Generic composition that lacks the requested surrealism.
Verdict: FLUX.1 [schnell] FP8 successfully followed the difficult spatial instruction of having the horse on top, resulting in a truly surreal and cinematic image set in space. HiDream I1 Fast failed almost every part of the prompt, delivering a standard astronaut riding a horse in a desert, which is the direct opposite of what was requested.
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 capybara paws on the steering wheel as requested
- + Detailed New York street background through the window
- + Realistic lighting and shadow integration within the taxi
- − The passenger is holding two phones simultaneously
- − The capybara's hat looks a bit more like a plastic toy than a fabric driver cap
HiDream I1 Fast
- + Realistic interpretation of a driver's cap
- + Clear bokeh effect on the background city lights
- − The capybara has human hands, which is a significant anatomical error
- − The passenger is sitting in the front passenger seat rather than the back seat
- − The capybara is not looking forward but staring straight into the door frame
Verdict: FLUX.1 [schnell] FP8 followed the prompt instructions much better, correctly placing the passenger in the back and giving the capybara paws rather than human hands. While HiDream I1 Fast has nice lighting, the anatomical failure of giving the animal human hands and the incorrect seat placement for the passenger makes it a much weaker image.
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
- + Deep, cinematic lighting and high-quality textures
- + Excellent atmospheric rendering of the jack-o-lantern and bats
- + Effective use of the square format with a moody, dark aesthetic
- − Significant text hallucinations and spelling errors in the footer details
- − The banner text is split awkwardly into two separate banners
HiDream I1 Fast
- + Accurate date rendering (30.10.2026)
- + Clearly defined gothic visual elements like the webs and torn parchment
- + Stronger adherence to the 'vintage' aspect of the prompt
- − Text in the scroll banner is cut off and contains typos
- − Address and time details are repetitive and messy
- − Visual style leans more towards a cartoon illustration than cinematic
Verdict: FLUX.1 [schnell] FP8 delivers a much more polished and atmospheric image with superior lighting, but fails significantly on text legibility. HiDream I1 Fast handles the specific date and vintage prompt elements better, but the overall image quality is lower with distracting text artifacts in the footer. FLUX.1 [schnell] FP8 is the preferred choice for its visual impact, despite the nonsensical 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 [schnell] FP8
- + Excellent soft 3D textures and lighting
- + Matches the solid light blue background color perfectly
- + Clean isometric perspective on a diorama base
- − Failed text rendering, missing 'SUSHI' and misspelled 'JAPAN'
- − Distorted flag icon integrated poorly into the text
HiDream I1 Fast
- + Accurate text rendering for both 'JAPAN' and 'SUSHI'
- + Includes appropriate flag icon as requested
- + Good 3D miniature style with pleasing colors
- − Lower visual clarity compared to the competitor
- − Text layout is a bit cramped with the horizontal lines
- − Foreground chopsticks were not requested and add clutter
Verdict: HiDream I1 Fast is the clear winner for prompt adherence as it successfully rendered the requested text and flag icon, which FLUX.1 [schnell] FP8 failed to do. While FLUX.1 [schnell] FP8 had slightly more refined 3D textures, its failure to generate readable text and its misspelling of the primary subject make it less effective for this specific prompt.
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
- + Includes a high number of butterflies and dynamic poses
- + Strong lighting effects with visible god rays
- − Anatomy failures include a two-headed animal in the bottom right and extra limbs
- − Failed to include the rabbit
- − Style is overly 'digital' and lacks the requested photorealistic quality
HiDream I1 Fast
- + Excellent fur texture and realistic anatomy for all subjects
- + Closer to the requested photorealistic style
- + Warm, balanced composition with soft lighting
- − Failed to include the rabbit
- − Butterflies appear slightly pasted on with inconsistent lighting
Verdict: While both models failed to include the requested baby bunny, HiDream I1 Fast is the superior image due to its 훨씬 more realistic anatomy and fur textures. FLUX.1 [schnell] FP8 suffers from significant AI artifacts, including a merged/two-headed creature in the foreground and a more cartoonish aesthetic.
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
- + Successfully captured the requested color palette and texture.
- + Excellent vector-style line art and composition.
- + Includes the correct establishment year in the requested location.
- − Serious spelling errors in the brand name, rendering 'AFe FLAMILAN'.
- − The cloche dome is stylized to look more like a building dome or coffee pot top.
HiDream I1 Fast
- + Accurately spelled the brand name 'CAFFÈ FLORIAN'.
- + Clearly depicts a food cloche dome as requested in the prompt.
- + Pleasing vector aesthetic with balanced symmetry.
- − Typo in the date, displaying '17210' instead of '1720'.
- − The steam effect is a bit clunky and less integrated than in the other model.
Verdict: HiDream I1 Fast is the clear winner for its superior text rendering of the restaurant name, whereas FLUX.1 [schnell] FP8 completely fails to spell 'Florian'. While HiDream I1 Fast has a minor typo in the date ('17210'), it accurately portrays the cloche dome requested by the prompt, which FLUX.1 [schnell] FP8 interpreted more as generic architecture.
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 navy, white, and muted red color palette.
- + Crisp, modern vector aesthetic with consistent iconography.
- + Good spatial layout that follows a logical chronological timeline.
- − Text rendering is largely gibberish despite attempting labels.
- − The icons do not perfectly match the specific Saturn V or Lunar Module descriptions.
HiDream I1 Fast
- + Successfully renders the Saturn V rocket central figure.
- + Includes a recognizable US flag and landing site imagery.
- + Readable main headings for each step.
- − Layout is cluttered and lacks a clear chronological flow.
- − Text rendering features multiple spelling errors like 'DESCENG' and '8bLt lrite'.
- − Visual style is less 'clean and modern' compared to the requested flat-vector look.
Verdict: FLUX.1 [schnell] FP8 much better captures the requested professional 'modern vector infographic' aesthetic and NASA-inspired color palette, presenting a clean and organized layout. While HiDream I1 Fast succeeds in rendering more literal interpretations of the rocket and flag, its composition is messy and the 'flat-vector' style is less refined. FLUX.1's superior design quality makes it the winner despite the nonsensical sub-text.
Explore each model
Distilled version of HiDream AI's 17B parameter text-to-image model