Black Forest Labs' open-weights multimodal flow transformer for in-context image generation and editing, available for non-commercial use with character consistency and style transfer capabilities
Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.
FLUX.1 Kontext [dev]
17.2
arena score
#54 of 62 in Text-to-Image
Skill signature
· Text-to-Image
FLUX.1 [schnell] FP8
19.0
arena score
#47 of 62 in Text-to-Image
Vote tally
Where the votes landed
FLUX.1 Kontext [dev]
0%
win rate
Ties
0%
FLUX.1 [schnell] FP8
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 Kontext [dev]
FLUX.1 [schnell] FP8
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Excellent adherence to the cube geometry.
- + Highly realistic textures on the red book and the wooden table.
- + Accurate lighting and reflections on the glass and blue sphere.
- − The plant visibility through the glass is slightly subtle due to the mirror-like floor of the cube.
FLUX.1 [schnell] FP8
- + Beautiful translucent effect on the blue sphere.
- + Good spatial arrangement of the elements.
- − The glass object is a rectangular prism rather than the requested cube.
- − The interior has an illogical horizontal shelf that was not in the prompt.
- − The perspective of the cube's base is slightly warped.
Verdict: FLUX.1 Kontext [dev] followed all spatial and geometric instructions, producing a perfect cube with a solid blue sphere as requested. FLUX.1 [schnell] FP8 failed to generate a cube, instead creating a tall rectangular column with an added internal shelf, which deviated from the prompt.
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 Kontext [dev]
FLUX.1 [schnell] FP8
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Strong adherence to the rain and reflection prompts
- + High level of detail in the bicycle mechanics and tires
- + Good cinematic lighting from the car headlights
- − The man is posing with the bike rather than repairing it
- − Lack of requested motion blur for passing cars
- − The rain looks like a static overlay rather than part of the environment
FLUX.1 [schnell] FP8
- + Successfully captures the action of repairing or adjusting the bike
- + Better use of 'imperfect framing' and candid perspective
- + More realistic skin texture and natural lighting
- − Missed the 'passing car motion blur' requirement as cars are stationary
- − The man appears less specifically Japanese compared to Model A
- − Rain is hardly visible except for the wet ground
Verdict: FLUX.1 Kontext [dev] creates a more visually striking 'cinematic' image with clear rain and reflections, but the subject is simply standing mid-street rather than repairing the bike. FLUX.1 [schnell] FP8 captures the 'candid' and 'repairing' aspect much better with more natural posing and framing, though it fails to clearly depict the falling rain requested.
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 Kontext [dev]
FLUX.1 [schnell] FP8
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Excellent depiction of ornate engraved plate armor with clear patterns.
- + Strong composition that balances the figure with atmospheric bokeh sparks.
- + Realistic scars and dirt that convey the battle-worn theme.
- − Hair lacks the explicit braiding and bead details requested.
- − Metallic textures look slightly flat compared to the lighting expectations.
FLUX.1 [schnell] FP8
- + Successfully includes the requested beads in the hair.
- + Extreme close-up provides incredible skin texture and iris detail.
- + Dynamic lighting with strong highlights and shadows across the face.
- − Armor is largely cropped out, missing the 'ornate engraved plate' focus.
- − The expression is somewhat overly stylized/contrasty compared to a natural portrait.
- − Includes nonsensical text artifacts on the armor at the bottom right.
Verdict: Both models captured the mood well, but FLUX.1 Kontext [dev] is the superior choice for this specific prompt as it correctly prioritizes the engraved plate armor and a full portrait composition. While FLUX.1 [schnell] FP8 has more impressive fine skin detail and includes the hair beads, it fails to show the requested armor significantly and introduces AI 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.”
FLUX.1 Kontext [dev]
FLUX.1 [schnell] FP8
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Strong, high-quality food photography with vibrant colors
- + Bold, modern typography that fits the minimalist aesthetic
- + Creative checkerboard grid layout that feels contemporary
- − Nonsense text is very prominent and garbled
- − Layout is more like a poster or social media graphic than a functional multi-page menu
FLUX.1 [schnell] FP8
- + Excellent structural adherence to a multi-section restaurant menu
- + Includes specific sections for pizza and appetizers as requested
- + Clean, professional layout that logicially presents prices and descriptions
- − Food photos are repetitive and lack the high-end appeal of model a
- − Text becomes unreadable at smaller sizes
Verdict: FLUX.1 Kontext [dev] produces an image with superior visual quality and photography, but it fails to create a functional menu structure. FLUX.1 [schnell] FP8 successfully interprets the 'restaurant menu design' prompt by including multiple sections, pricing, and a logical reading flow, making it the more useful interpretation of the design brief.
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 Kontext [dev]
FLUX.1 [schnell] FP8
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Stronger fiery aesthetic with glowing embers that match the prompt atmosphere.
- + Clean typography for 'MAGIC BURGER' and '€6.99'.
- + Excellent photorealistic details in the lettuce and toasted bun.
- − Failed the 'exploded burger' requirement, showing a fully assembled burger instead.
- − Typo in the secondary text: 'INLLY' instead of 'ONLY'.
FLUX.1 [schnell] FP8
- + Successfully captured the 'exploded' effect with suspended components.
- + High degree of dynamic motion in the composition.
- + Includes the requested starburst for the price.
- − Multiple severe text errors including 'LIIMITED', 'LIMIED', and 'NEEY'.
- − Incorrect price rendering (€69 and €.99) instead of €6.99.
- − The overall image quality feels slightly more synthetic compared to Model A.
Verdict: The comparison is a trade-off between conceptual adherence and technical polish. FLUX.1 [schnell] FP8 correctly followed the 'exploded burger' layout but failed significantly on text accuracy and price rendering, whereas FLUX.1 Kontext [dev] produced a much higher quality image with better atmosphere but completely ignored the request for an exploded view.
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 Kontext [dev]
FLUX.1 [schnell] FP8
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Excellent chalk texture and realistic handwriting variations.
- + The header and dates are reasonably clear despite some spelling artifacts.
- + The overall composition feels authentic to a café atmosphere.
- − Contains significant spelling errors like 'Mashroom', 'Risoktso', and 'Octpus'.
- − Repeats words and prices within the same line, causing clutter.
FLUX.1 [schnell] FP8
- + Successfully renders the date 'APRIL 30, 2026' very clearly.
- + Good lighting and frame depth within the café setting.
- − Severe spelling hallucinations and garbled text across all menu items.
- − The cursive requirement for the title was not fully met, appearing more as standard print.
- − The text becomes nonsensical gibberish toward the bottom of the board.
Verdict: Both models struggled with the complex spelling requirements of the menu items, but FLUX.1 Kontext [dev] produced a far more convincing chalkboard aesthetic with superior texture and font variation. While FLUX.1 [schnell] FP8 rendered the date correctly, its breakdown into gibberish for the menu items makes it less usable than FLUX.1 Kontext [dev].
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
FLUX.1 Kontext [dev]
FLUX.1 [schnell] FP8
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Excellent adherence to the 'horse on top' spatial instruction
- + Clear and coherent anatomical rendering of both subjects
- + High cinematic quality with professional lighting
- − The horse appears to be standing behind rather than literally riding the astronaut
- − Relatively simple background
FLUX.1 [schnell] FP8
- + Dynamic composition with a large earth backdrop
- + Strong surreal aesthetic
- − Failed the core prompt instruction of 'horse on top', featuring horses riding machinery instead
- − Extreme anatomical distortions with horse limbs blending into equipment
- − Nonsensical elements like a roof at the bottom of the frame
Verdict: FLUX.1 Kontext [dev] followed the difficult spatial instruction much better, correctly placing the horse above/on the astronaut, whereas FLUX.1 [schnell] FP8 produced a chaotic image of horses riding pieces of a spacecraft. FLUX.1 Kontext [dev] is the clear winner for its coherence and prompt adherence, despite the horse appearing more adjacent than directly seated on the astronaut.
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 Kontext [dev]
FLUX.1 [schnell] FP8
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Features more realistic, cinematic lighting and color grading
- + Capybara's expression is very calm and fits the prompt well
- + Better attention to fur texture and realistic anatomy for the capybara
- − One paw is resting on the lap instead of 'both front paws on the steering wheel'
- − The woman's expression looks slightly distressed rather than bored
FLUX.1 [schnell] FP8
- + Successfully places both paws on the steering wheel
- + Includes clear 'TAXI' text on the cap
- + The woman's expression is more bored and nonchalant as requested
- − The woman is holding two phones, which is a logic error
- − Composition is a bit crowded and lighting feels more artificial/flat
- − The capybara's anatomy is slightly more humanoid/less animal-accurate
Verdict: FLUX.1 Kontext [dev] produces a much more convincingly photorealistic image with superior lighting and texture, though it misses the minor instruction of keeping both paws on the wheel. FLUX.1 [schnell] FP8 follows the composition and character instructions more literalistically but suffers from a significant error where the woman is holding two phones, and the overall image quality is less refined.
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 Kontext [dev]
FLUX.1 [schnell] FP8
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Excellent large title text rendering
- + Strong jack-o-lantern glow and details
- + Followed the thorny border requirement well
- − Banner text is messy and illegible
- − Venue location 'The Arches' is misspelled as 'ARGIIVAH'S'
- − Lacks the 'moody night sky' element requested
FLUX.1 [schnell] FP8
- + Atmospheric moody night sky with cinematic lighting
- + Better scroll banner aesthetics
- + More interesting vintage parchment texture
- − Major spelling and data errors in the event details section
- − Missed the specific word 'Party' in the main title
- − Jack-o-lantern looks somewhat artificial and plastic-like
Verdict: FLUX.1 Kontext [dev] produced a much clearer and more readable main title and better iconography, but struggled significantly with the venue name and secondary text. FLUX.1 [schnell] FP8 captured the atmospheric and 'moody' prompt requirements more effectively but failed largely on text accuracy and data placement. Overall FLUX.1 Kontext [dev] is the better choice for an invitation due to its superior central graphic and main heading legibility.
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 Kontext [dev]
FLUX.1 [schnell] FP8
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Excellent typography rendering with bold, readable text.
- + Very clean, minimalist aesthetic that fits the 'cartoon' and 'diorama' request well.
- + Successful 45-degree isometric composition.
- − The flag icon is abstract and does not accurately represent the Japanese flag.
- − The single piece of sushi is overly simplified, losing some of the 'signature dish' feel.
FLUX.1 [schnell] FP8
- + Features a more variety-rich and visually interesting selection of sushi.
- + Beautiful soft lighting and refined, clay-like textures.
- + The flag icon is more recognizable as the Japanese flag.
- − Failed the text prompt, repeating 'JAPAN' incorrectly as 'JAPAN JA-AN' instead of including 'SUSHI'.
- − Low contrast between white text and light blue background makes it hard to read.
Verdict: FLUX.1 Kontext [dev] followed the text instructions perfectly, providing clear and legible typography, though the central model is slightly too simplified. FLUX.1 [schnell] FP8 produced a more charming and detailed 3D diorama of the sushi itself, but failed significantly on the specific text requirements and name placement. FLUX.1 Kontext [dev] is the winner for total prompt adherence and clarity.
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 Kontext [dev]
FLUX.1 [schnell] FP8
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Excellent action-oriented composition that shows animals leaping and chasing butterflies.
- + Clean, high-quality rendering of fur and lighting.
- + Anatomically coherent figures with distinct species traits.
- − Failed to include all four requested animals, missing the baby bunny and red fox kit.
- − Atmosphere is slightly more generic with less prominent god rays.
FLUX.1 [schnell] FP8
- + Includes a larger variety of animals, attempting several different species from the prompt.
- + Stunning use of golden sunrise light and clearly visible god rays.
- + Captures the 'tumbling together' aspect with close physical interaction between the animals.
- − Anatomical errors such as a kitten having too many legs or paws merging with others.
- − Some animals appear as hybrids rather than distinct species (e.g., the fox/kitten blends).
- − Lack of a clear baby bunny as requested.
Verdict: FLUX.1 Kontext [dev] produced a much cleaner and more realistic image, but it failed significantly on prompt adherence by only including three animals. FLUX.1 [schnell] FP8 captured the specific lighting and 'tumbling' interaction better and included more animals, but it suffered from logical and anatomical errors in the fur and limbs. FLUX.1 Kontext [dev] is the winner for its superior visual quality and lack of disturbing artifacts, despite missing two of the requested animals.
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 Kontext [dev]
FLUX.1 [schnell] FP8
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Perfect text rendering of the restaurant name and year
- + Clean, minimalist vector aesthetic consistent with modern logo design
- + Excellent clarity and high-contrast color palette
- − Missed the 'banner' element from the prompt
- − Typography is a bit too modern/bold for a truly 'vintage' feel
FLUX.1 [schnell] FP8
- + Better 'vintage' aesthetic with classic serif typography and decorative flourishes
- + Successfully included the requested banner element
- + Good adherence to the 'cloche dome' and 'steam' visual motifs
- − Severely failed the text rendering, misspelling the name as 'Afe Flamilan' and 'Caffeé'
- − Poor composition with text overlapping and clipping through the banner
Verdict: FLUX.1 Kontext [dev] followed the text requirements perfectly, producing a crisp, professional logo, though it lacked the requested banner. FLUX.1 [schnell] FP8 captured the vintage style and the banner element much better, but completely failed to render the text correctly, resulting in a nonsensical name. FLUX.1 Kontext [dev] is the clear winner for its legibility and high-quality vector execution.
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 Kontext [dev]
FLUX.1 [schnell] FP8
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Strong adherence to the requested color palette
- + Good vector-style line work on icons
- − Major spelling errors in the main title ('Apolo')
- − The layout is cluttered and the text overlays are garbled and illegible
- − The icons are abstract and do not clearly represent the requested mission steps
FLUX.1 [schnell] FP8
- + Excellent logical layout with a clear horizontal timeline
- + Accurate spelling of 'Apollo 11' in the header
- + Better adherence to the requested iconography including the trajectory arc and lunar module
- − The placeholder body text is gibberish
- − The color palette is slightly less vibrant than requested, though still relevant
Verdict: FLUX.1 [schnell] FP8 is the clear winner as it successfully interprets the request for an infographic through a structured timeline layout and legible main titles. While FLUX.1 Kontext [dev] has a nice aesthetic, it fails on basic spelling for the mission name and the layout is too chaotic to function as an educational poster.
Next steps
Explore each model
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