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
19.0
arena score
#47 of 62 in Text-to-Image
Skill signature
· Text-to-Image
GPT Image 1 Mini
25.0
arena score
#13 of 62 in Text-to-Image
Vote tally
Where the votes landed
FLUX.1 [schnell] FP8
0%
win rate
Ties
0%
GPT Image 1 Mini
0%
win rate
Shared challenges
13
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.”
FLUX.1 [schnell] FP8
GPT Image 1 Mini
AI Judge Analysis
FLUX.1 [schnell] FP8
- + High visual clarity and realistic reflections
- + Artistic interpretation of the blue sphere with internal detail
- + Excellent lighting and background depth
- − The glass object is a tall rectangular prism rather than a cube
- − The sphere appears to be floating on an internal shelf rather than sitting on the base
GPT Image 1 Mini
- + Perfect adherence to the geometry of a cube
- + Accurate placement of all requested elements
- + Natural-looking wooden table texture
- − Lighting is somewhat flat compared to the other model
- − The plant lacks the detail and translucency of the other model
Verdict: While FLUX.1 [schnell] FP8 produces a more visually stunning image with beautiful light play, it fails on basic geometry by creating a rectangle instead of a cube. GPT Image 1 Mini followed the prompt instructions precisely, capturing the cube shape and the specific arrangement of objects perfectly, making it the better choice for prompt adherence.
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.”
FLUX.1 [schnell] FP8
GPT Image 1 Mini
AI Judge Analysis
FLUX.1 [schnell] FP8
- + Excellent depiction of wet pavement and city light reflections
- + Good camera lens simulation with 50mm feel and shallow depth of field
- + Strong cinematic color palette
- − Failed to include 'light rain' visual effects like droplets or mist
- − Missing the requested 'motion blur from passing cars' as the background cars are sharp
- − The subject is just standing with the bike rather than 'repairing' it
GPT Image 1 Mini
- + Superior adherence to the specific action of 'repairing' the bike while crouching
- + Excellent rendition of skin texture and elderly facial features
- + Visible rain droplets on the bike seat and atmosphere capture the 'light rain' prompt well
- − Fails to show 'reflections on wet pavement' due to the tight framing
- − The background car lacks 'motion blur' requested in the prompt
Verdict: GPT Image 1 Mini captures the candid nature and specific action of the prompt much better, providing realistic skin textures and the subtle detail of rain droplets on surfaces. While FLUX.1 [schnell] FP8 handles environmental lighting and wet pavement better, it misses the core 'repairing' action and the presence of rain, resulting in a more generic portrait.
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.”
FLUX.1 [schnell] FP8
GPT Image 1 Mini
AI Judge Analysis
FLUX.1 [schnell] FP8
- + Strong, dramatic lighting and high-contrast skin textures
- + Intense, lifelike eye rendering
- + Exceptional sharpness and clarity in the foreground
- − Missed the request for braided hair with beads
- − The armor texture is somewhat obscured by the tight framing
- − Skin looks overly saturated and slightly digital in some areas
GPT Image 1 Mini
- + Excellent adherence to the 'braided hair' and 'ornate engraved plate armor' prompts
- + Realistic battle-worn appearance with convincing dirt and faint scars
- + Beautiful bokeh sparks and atmospheric warm lighting
- − The eyes are a bit less expressive and detailed than Image A
- − The overall image is slightly softer and warmer, losing some fine skin pore detail
Verdict: GPT Image 1 Mini followed the prompt much more accurately, successfully incorporating the braided hair, beads, and heavy armor engravings that FLUX.1 [schnell] FP8 missed. While FLUX.1 produced a more striking and intense close-up portrait, GPT Image 1 Mini captured the specified 'battle-worn' aesthetic and armor details better.
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.”
FLUX.1 [schnell] FP8
GPT Image 1 Mini
AI Judge Analysis
FLUX.1 [schnell] FP8
- + Comprehensive menu layout with prices and descriptions
- + Included all requested sections and multiple food images
- + Professional use of white space and typography hierarchy
- − Nonsense text for many section headers and items
- − Some food images look slightly artificial or repetitive
GPT Image 1 Mini
- + Perfect text rendering for headers
- + Crips and high-quality food photography
- + Clean minimalist grid layout
- − Missing all menu items, descriptions, and prices
- − The left side of the layout is unnecessarily empty
- − Very basic design that lacks the 'professional' feel of a real menu
Verdict: FLUX.1 [schnell] FP8 creates a much more realistic menu template with sections, prices, and a double-page spread, though it struggles with spelling. GPT Image 1 Mini has perfect spelling and higher quality food photography, but fails to actually include the 'menu' content, leaving a large portion of the layout empty. FLUX.1 [schnell] FP8 is preferred as it better represents a functional restaurant menu design.
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.”
FLUX.1 [schnell] FP8
GPT Image 1 Mini
AI Judge Analysis
FLUX.1 [schnell] FP8
- + Excellent photorealistic texture of the burger ingredients
- + Vibrant and high-contrast color palette
- − Serious spelling errors in the 'Limited Time' text
- − Price is incorrect and poorly rendered in the starburst
- − The burger is mostly intact rather than 'exploded' with suspended components
GPT Image 1 Mini
- + Perfect adherence to all text requirements including '€6.99'
- + Captures the 'exploded' deconstructed view accurately
- + Excellent execution of the fiery glowing effect on typography
- − Food textures are slightly less sharp than Model A
- − Composition feels a bit more static compared to the debris in Model A
Verdict: GPT Image 1 Mini is the clear winner as it followed every instruction, including the specific pricing, the fiery text effect, and the exploded burger structure. FLUX.1 [schnell] FP8 failed significantly on the text rendering, providing multiple misspelled iterations and the wrong price.
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.”
FLUX.1 [schnell] FP8
GPT Image 1 Mini
AI Judge Analysis
FLUX.1 [schnell] FP8
- + Excellent realistic chalk texture on a blackboard with smudges and highlights.
- + Natural, high-contrast handwriting style.
- − Poor spelling and word repetition errors in the menu items.
- − Title is in block capitals rather than the requested elegant cursive.
GPT Image 1 Mini
- + Perfect text accuracy and spelling for all requested menu items.
- + Very convincing chalk-on-blackboard texture with realistic grain.
- + Includes a logical completion for the truncated prompt 'Brown But...'.
- − The handwriting style is a bit too uniform, bordering on a digital font look in certain sections.
- − Missed the request for 'elegant cursive' for the title.
Verdict: GPT Image 1 Mini is the clear winner due to its superior text rendering and adherence to the specific menu items requested. While FLUX.1 [schnell] FP8 captured a more artistic and messy 'handwritten' feel, its severe spelling errors and repetitive text ('Truffle Mushroom / ffle Risortto') make the menu unreadable, whereas GPT Image 1 Mini delivered a professional and accurate result.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
FLUX.1 [schnell] FP8
GPT Image 1 Mini
AI Judge Analysis
FLUX.1 [schnell] FP8
- + Excellent adherence to the 'horse on top' spatial instruction
- + Surreal and creatively interprets the 'astronaut' as tech equipment for the horse
- + Dynamic composition with a striking planetary backdrop
- − The anatomy of the two horse torsos merging into one central machine is a bit messy
- − A bit chaotic in its representation of what the 'astronaut' part is
GPT Image 1 Mini
- + High visual quality with realistic lighting and texture
- + Very cinematic and polished aesthetic
- + Anatomy of both the horse and the astronaut is technically perfect
- − Failed the core prompt instruction of 'horse on top, not vice versa'
- − Generic interpretation of a common surrealist trope
Verdict: While GPT Image 1 Mini produced a much more beautiful and professionally rendered image, it completely ignored the specific spatial instruction to place the horse on top of the astronaut. FLUX.1 [schnell] FP8 correctly interpreted the difficult prompt, creating a surreal image where a horse rides a technical astronaut-eqsue device, making it the winner for prompt adherence despite slightly messy details.
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.”
FLUX.1 [schnell] FP8
GPT Image 1 Mini
AI Judge Analysis
FLUX.1 [schnell] FP8
- + Excellent texture on capybaras fur
- + Captures the New York taxi aesthetic clearly even in the background
- + Modern, crisp visual style
- − The passenger is holding two phones unnaturally
- − The capybara's hat looks a bit more like a toy than a professional driver's cap
- − The composition feels a bit cramped due to the tight framing
GPT Image 1 Mini
- + Perfect adherence to the 'bored expression' of the passenger
- + More realistic clothing and accessories like the driver's hat with checkerboard pattern
- + Excellent lighting and cinematic atmosphere that feels more photorealistic
- − The passenger's hand on the phone is slightly blurry
- − One of the capybara's paws is not visible on the wheel as requested
Verdict: GPT Image 1 Mini captures the requested atmosphere and character expressions much more effectively, particularly the 'bored' passenger. FLUX.1 [schnell] FP8 delivers a high-quality image but suffers from logic errors like the passenger holding two smartphones.
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.”
FLUX.1 [schnell] FP8
GPT Image 1 Mini
AI Judge Analysis
FLUX.1 [schnell] FP8
- + Strong cinematic lighting on the jack-o-lantern
- + Vibrant colors and high contrast
- − Significant spelling errors throughout the invitation text
- − Muddled border that looks like torn leather rather than gothic thorns/webs
GPT Image 1 Mini
- + Excellent typography with nearly perfect spelling
- + Captures the vintage parchment aesthetic much better
- + Intricate spiderweb and thorn border as requested
- − Lighting is a bit flatter compared to Image A
- − Slightly grainy texture across the entire image
Verdict: GPT Image 1 Mini is the clear winner because it successfully renders all the requested text with high accuracy and follows the 'vintage gothic' stylistic cues more closely. While FLUX.1 [schnell] FP8 has more dynamic lighting, its failure to spell the event details and invitation phrase makes it unusable as a functional poster.
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.”
FLUX.1 [schnell] FP8
GPT Image 1 Mini
AI Judge Analysis
FLUX.1 [schnell] FP8
- + Excellent soft lighting and PBR material simulation on the fish
- + Accurate 45-degree isometric perspective
- − Significant text failure, repeating 'JAPAN' and misspelling it twice
- − The background is more of a gradient/white than the requested solid light blue
GPT Image 1 Mini
- + Perfect adherence to text instructions including 'JAPAN', 'SUSHI', and the flag icon
- + High visual clarity with clean, well-defined textures
- + Strong composition on a clear diorama base
- − The base textures look slightly more like plastic than a 'refined diorama'
Verdict: GPT Image 1 Mini followed the complex text-rendering and layout instructions perfectly, whereas FLUX.1 [schnell] FP8 failed significantly on the typography. While FLUX.1 had slightly more realistic sub-surface scattering on the salmon, GPT Image 1 Mini is the clear winner for its overall adherence to every prompt constraint including the specific labels.
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.”
FLUX.1 [schnell] FP8
GPT Image 1 Mini
AI Judge Analysis
FLUX.1 [schnell] FP8
- + Extremely high level of fur detail and lighting vibrance
- + Excellent expressive 'big eyes' as requested in the prompt
- + Whimsical atmosphere with beautiful butterfly placement and god rays
- − Failed to include a rabbit, instead showing five cat/fox hybrids
- − Morphology of the animals is inconsistent, resulting in strange hybrid creatures
- − More illustrative than 'hyper-photorealistic'
GPT Image 1 Mini
- + Successfully included all four distinct animal species: puppy, kitten, bunny, and fox
- + Captured the 'playfully chasing' action much better with dynamic poses
- + Strong adherence to the 'photorealistic' style with natural anatomy
- − Lighting is slightly flatter compared to Model A
- − Fewer butterflies and less 'glow' than the prompt suggested
- − The kitten's tail looks slightly disconnected from its body
Verdict: GPT Image 1 Mini is the clear winner because it correctly followed the instructions to include a puppy, kitten, bunny, and fox, whereas FLUX.1 [schnell] FP8 failed to generate a bunny and instead created multiple similar-looking fox/cat hybrids. GPT Image 1 Mini also better captured the 'chasing' and 'tumbling' action from the prompt, maintaining much more realistic animal anatomy.
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.”
FLUX.1 [schnell] FP8
GPT Image 1 Mini
AI Judge Analysis
FLUX.1 [schnell] FP8
- + Excellent vintage color palette with warm brown and cream tones
- + Intricate vector emblem aesthetic
- + Accurate texture on a light background
- − Significant spelling errors in the main brand name
- − The graphic looks more like an architectural dome than a food cloche
GPT Image 1 Mini
- + Perfect text rendering for both the brand name and the date
- + Clearly identifiable cloche dome with steam
- + Strong minimalist composition
- − Failed the request for a light background, providing black instead
- − Subtle texture is less visible compared to the other model
Verdict: GPT Image 1 Mini correctly identifies all elements of the prompt including the cloche and perfect typography, but it completely ignored the instruction for a light background. FLUX.1 [schnell] FP8 captured the requested artistic style and color palette much better, but failed significantly on spelling and character recognition.
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.”
FLUX.1 [schnell] FP8
GPT Image 1 Mini
AI Judge Analysis
FLUX.1 [schnell] FP8
- + Stronger vector aesthetic and modern layout
- + Excellent use of the requested navy background and muted palette
- + Clean typography for the main title
- − Nonsense filler text for all descriptions
- − Icons are abstract and don't match the specific prompt descriptions (e.g., no Saturn V icon)
- − Visual information is confusingly laid out
GPT Image 1 Mini
- + Highly accurate iconography following each step of the prompt
- + Excellent legibility and text rendering
- + Clean, consistent illustrative style that conveys the mission steps clearly
- − Off-white background instead of the requested navy theme
- − The 'Translunar' trajectory line is visually messy and lacks a clear 'icon'
Verdict: GPT Image 1 Mini followed the instructional steps of the prompt much more accurately, providing identifiable icons for the Saturn V and Lunar Module with clear, legible text. While FLUX.1 [schnell] FP8 captured the 'modern vector infographic' atmosphere and navy color scheme better, it failed to generate recognizable icons or meaningful text for the specific mission steps.
Next steps
Explore each model
OpenAI's cost-effective image generation model for when image quality isn't the top priority