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 17 shared challenges, with an AI judge weighing in on each.
FLUX.1 Kontext [dev]
#54 of 62 in Text-to-Image
FLUX.1 Kontext [max]
#23 of 62 in Text-to-Image
Where the votes landed
FLUX.1 Kontext [dev]
0%
win rate
Ties
0%
FLUX.1 Kontext [max]
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 Kontext [dev]
- + Excellent literal interpretation of the green plant behind the glass
- + Superior material rendering for the glass and reflected light
- + The sphere appears more 'inside' the cube due to the tight framing
- − The sphere is quite large relative to the term 'small' in the prompt
- − The book scale is a bit large for the cube
FLUX.1 Kontext [max]
- + Better adherence to the 'small' sphere instruction
- + More dramatic and realistic window lighting effects on the table surface
- + Better scale relationship between the object and the environment
- − The plant is more behind the book than visible 'through' the glass as requested
- − The sphere texture looks slightly fuzzy/glittery rather than smooth
Verdict: Both models followed the complex spatial instructions perfectly. FLUX.1 Kontext [dev] is preferred for his superior rendering of transparency, specifically showing the plant through the glass, while FLUX.1 Kontext [max] produced much better lighting aesthetics and more accurate scale for the 'small blue sphere'.
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 Kontext [dev]
- + Features excellent reflections on the wet pavement
- + Good quality and sharp focus on the subject
- − The subject is posing with the bike rather than 'repairing' it as requested
- − The bicycle has structural issues, specifically the seat and rear wheel alignment
- − Misses the 'motion blur from passing cars' request
FLUX.1 Kontext [max]
- + Strong adherence to the 'repairing' action and 'imperfect framing' prompt
- + Better skin texture and realistic hand lighting
- + Captures a more cinematic and candid street photography feel
- − The rain effect looks like a static layer of vertical white lines
- − The bicycle chain and wheel spokes are somewhat jumbled in the foreground
Verdict: FLUX.1 Kontext [max] is the winner because it successfully captured the action of 'repairing' and achieved a convincing street photography aesthetic with imperfect framing. While FLUX.1 Kontext [dev] produced a cleaner image with better reflections, the subject failed to perform the requested action, appearing to just be standing with a broken bicycle in the middle of a road.
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 Kontext [dev]
- + Excellent engraving detail on the plate armor
- + Clean, high-quality lighting that creates a dramatic mood
- + Strong adherence to the shallow depth of field request
- − Missed the request for braided hair with beads
- − The skin lacks the requested level of dirt and 'battle-worn' texture, appearing quite clean
FLUX.1 Kontext [max]
- + Superb skin texture showing sweat, grime, and realistic pores
- + Successfully included braids in the hair as requested
- + Excellent rendering of lifelike, piercing eyes
- − The sparks look a bit like digital streaks rather than natural embers
- − The braids appearing over the front of the armor looks slightly unnatural in composition
Verdict: FLUX.1 Kontext [max] is the winner as it adhered to significantly more of the prompt requirements, specifically the braided hair and the detailed skin textures including dirt and grime. While FLUX.1 Kontext [dev] produced a very aesthetically pleasing image with superior armor engravings, it failed to reflect the 'battle-worn' and 'braided' aspects of the character description.
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 Kontext [dev]
- + High-resolution, detailed food photography
- + Clean minimalist layout with high contrast
- + Bold, readable sans-serif typography
- − Nonsense text throughout the layout
- − Composition feels more like a magazine spread than a menu
FLUX.1 Kontext [max]
- + Functional menu structure with pricing and sections
- + Clear brand identity with a logo and cohesive color palette
- + Accurate representation of a professional casual dining menu
- − Food photos are repetitive (mostly pizza)
- − Text becomes garbled and messy at smaller scales
- − Lower visual clarity in food textures compared to the competitor
Verdict: FLUX.1 Kontext [max] succeeds by creating an actual menu structure that looks ready for use in a restaurant, complete with headers, prices, and a logical flow. While FLUX.1 Kontext [dev] has much higher quality photography and more striking typography, it fails to organize the content into a recognizable menu, appearing more like a portfolio or magazine page. FLUX.1 Kontext [max] is the preferred choice for actually meeting the functional requirements of the prompt.
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 Kontext [dev]
- + Excellent typography rendering with zero spelling errors
- + Clean and professional composition suitable for a social media ad
- + Sharp and photorealistic texture on the burger and coal
- − Failed to create an 'exploded' view; the burger is fully assembled
- − The starburst is a flat 2D graphic that clashes with the 3D scene
FLUX.1 Kontext [max]
- + Successfully interpreted 'exploded burger' with components flying apart
- + Dynamic sense of motion with particles and flying ingredients
- + Stronger aesthetic integration of text and background elements
- − The price text used a comma instead of a period
- − The starburst element is tiny and poorly defined
- − Lighting on the burger is slightly flatter compared to Image A
Verdict: FLUX.1 Kontext [max] is the winner for its better adherence to the 'exploded' and 'dynamic' parts of the prompt, creating a much more interesting visual story despite a punctuation error in the price. FLUX.1 Kontext [dev] produced a very clean and professional advertisement, but failed to separate the burger components as requested.
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 Kontext [dev]
- + Features a warm, centered composition for the chalkboard.
- + Achieves a high-quality chalk texture on the 'TODAY SPECIALS' text.
- − Numerous spelling errors including 'Mashroom', 'Risoktso', and 'Octpus'.
- − Randomly repeats words and numbers like 'with with' and '$28 - $28'.
- − Failed to render the month 'APRIL' correctly, using garbled characters instead.
FLUX.1 Kontext [max]
- + Excellent spelling accuracy across all requested menu items and the date.
- + Highly realistic chalkboard background with natural smudges and erasings.
- + Stronger environment context showing the cafe setting in the background.
- − The 'TODAY'S SPECIALS' text is in a blocky print rather than the requested elegant cursive.
- − The chalk texture on the main title looks a bit more uniform/digital compared to the body text.
Verdict: FLUX.1 Kontext [max] is the clear winner due to its superior text rendering and spelling accuracy, correctly identifying all menu items and the specific date requested. While FLUX.1 Kontext [dev] captured a nice chalk texture for the header, it failed significantly on prompt adherence with multiple typos and repetitive text artifacts.
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 Kontext [dev]
- + Successfully followed the specific instruction for the horse to be on top of the astronaut
- + Crisp rendering of the astronaut's face and suit textures
- + Conceptual interpretation of the prompt is creative
- − The composition is a bit rigid and staged
- − Visual coherence of the horse's positioning relative to the astronaut's back is slightly awkward
FLUX.1 Kontext [max]
- + High visual quality with beautiful lighting and star effects
- + Realistic rendering of the horse's mane and musculature
- − Failed the core prompt instruction of having the horse on top of the astronaut
- − Rendered a cliché scene that ignores the 'not vice versa' constraint
Verdict: FLUX.1 Kontext [dev] is the winner because it actually followed the specific and difficult instruction to have the horse on top of the astronaut. FLUX.1 Kontext [max] produced a higher quality image in terms of lighting and texture but completely failed the prompt by rendering a standard astronaut riding a horse.
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 Kontext [dev]
- + Captures a more professional 'driver' expression on the capybara.
- + Includes both figures clearly with the requested 'bored' expression for the passenger.
- + The texture of the capybara's fur and the jacket is highly realistic.
- − The passenger appears to be in the front-right seat rather than the back seat.
- − One paw is on the lap instead of both paws on the steering wheel.
- − The lighting inside the cab is quite dark, obscuring some interior details.
FLUX.1 Kontext [max]
- + Excellent photographic quality with vibrant, blurred city lights in the background.
- + Capybara is clearly wearing a 'TAXI' cap and has paws correctly positioned towards the wheel.
- + Composition feels more cinematic and dynamic.
- − The passenger is on the phone rather than looking at it, and her placement is ambiguous relative to the seats.
- − The capybara's head shape is slightly distorted by the hat placement.
Verdict: Both models struggled with the specific spatial arrangement of the passenger in the back seat, with both placing the human in what looks like the front passenger area. FLUX.1 Kontext [dev] captured the 'bored' expression of the businesswoman better, while FLUX.1 Kontext [max] produced a more visually striking image with superior lighting and a more legible taxi cap.
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 Kontext [dev]
- + Excellent readability for the main header text
- + The glowing jack-o-lantern has high contrast and vibrancy
- − The text inside the scroll banner is gibberish
- − The location text is misspelled as 'The Argiiah's'
- − Missing the dark parchment texture and spiderwebs
FLUX.1 Kontext [max]
- + Accurate rendering of almost all requested text including the banner
- + Beautiful dark parchment texture and intricate gothic border with webs
- + Strong cinematic lighting and atmosphere that matches the gothic theme
- − Uses a comma instead of a period in the date (30,10,2026)
- − The banner scroll ends are disconnected from the banner body
Verdict: FLUX.1 Kontext [max] is the clear winner as it successfully captures the vintage gothic aesthetic and dark parchment texture requested in the prompt. While FLUX.1 Kontext [dev] has clear headers, it fails significantly on the smaller text and misses the atmospheric 'parchment' and 'web' details that define the style.
Bald man challenge
Image Editing“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Successfully added a full head of hair that follows the direction of the prompt.
- + Maintained the overall color palette and lighting of the scene.
- − Significantly altered the facial features, making the person look younger and less weathered than the original.
- − The hairline and hair texture look slightly airbrushed and less natural than Model B.
FLUX.1 Kontext [max]
- + Excellent source preservation, keeping the age, wrinkles, and distinctive character of the original face.
- + Highly realistic hair texture with messy strands that blend better with the desert environment.
- + Maintained the original skin texture and facial anatomy.
- − The transition area where the hair meets the forehead shows a slight lighting mismatch near the center.
- − Minor artifacts where the hair meets the background bushes.
Verdict: While both models followed the instruction to add hair, FLUX.1 Kontext [max] is the winner because it successfully preserved the identity of the person in the source image. FLUX.1 Kontext [dev] beautified the subject too much, removing the aged details and 'weathered' look of the original man, whereas [max] kept the original face intact and provided a more realistic, messy hair texture appropriate for the setting.
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 Kontext [dev]
- + Excellent text rendering and alignment.
- + Clean and simple interpretation of the 'cartoon scene' with soft textures.
- + Strict adherence to the solid light blue background requirement.
- − The sushi looks a bit abstract and less recognizable as fish.
- − The flag icon is a stylized geometric shape rather than a recognizable Japan flag.
FLUX.1 Kontext [max]
- + Beautifully rendered miniature diorama with higher detail and better lighting.
- + Appealing materials that feel more like high-quality PBR assets.
- + Great composition and use of space for the isometric layout.
- − Failed to include the requested flag icon.
- − The text style is a bit more playful and less 'bold' than Model A.
Verdict: Both models followed the prompt well, but FLUX.1 Kontext [max] produced a much more visually appealing and detailed 'miniature diorama' as requested. While FLUX.1 Kontext [dev] had better text placement and included the flag icon, its sushi model felt overly simplified compared to the high-quality textures in the max version.
Over-the-top cartoon caricature
Editing“Create a caricature of me and my job. Make it exaggerated and humorous, incorporating my profession as a tv show anchor and my love for dogs and hockey.”
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Successfully adapts the woman's face into a comic-style caricature while maintaining her likeness from the source image
- + Includes a TV screen and dog to reference the profession and interests
- + Preserves the denim jacket and pose from the original photo
- − Completely misses the 'hockey' requirement of the prompt
- − Text in the image is nonsensical and distracting
- − The background elements feel a bit fragmented and low-effort
FLUX.1 Kontext [max]
- + Includes all three requested themes: TV news anchor, dog, and a hockey stick
- + The caricature style is very high quality with consistent lighting and texture
- + Excellent composition that clearly communicates the persona and setting
- − Likeness to the original woman is slightly weaker than Model A due to the addition of glasses not present in the source
- − The hockey stick is partially cut off at the top of the frame
Verdict: FLUX.1 Kontext [max] is the clear winner as it successfully incorporated all three specific thematic elements (TV anchor, dog, and hockey), whereas FLUX.1 Kontext [dev] failed to include the hockey theme entirely. FLUX.1 Kontext [max] also provides a more cohesive and professional-looking illustration, even if the addition of glasses makes the subject look slightly different from the source image.
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 Kontext [dev]
- + Dynamic action with the puppy jumping in the air
- + Clearer focus on individual characters
- + Good interaction with butterflies
- − Failed to include the bunny and the fox kit
- − Included two cats instead of the requested diverse mix
- − Lighting feels slightly more computer-generated and less natural
FLUX.1 Kontext [max]
- + Excellent prompt adherence including all four specific species
- + Captures the god rays and sunrise lighting effectively
- + Very soft, high-quality fur textures
- − Static composition compared to the 'chasing' prompt
- − The bunny has slightly unusual human-like eyes
Verdict: FLUX.1 Kontext [max] is the clear winner because it successfully generated all four requested animals (dog, cat, bunny, and fox), whereas FLUX.1 Kontext [dev] missed half of the subjects. While Kontext [dev] had more dynamic movement, Kontext [max] better captured the 'wholesome' atmosphere and complex subject list specified in the prompt.
Studio Ghibli Anime Style
Editing“Transform this photo into a Studio Ghibli–inspired illustration. Use soft pastel colors, hand-painted textures, gentle lighting, dreamy backgrounds, and a warm, nostalgic mood”
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Perfect preservation of the original meme composition and poses.
- + Clean anime-style cell shading.
- + Strong facial expressions that match the source image.
- − The style feels more like modern digital anime than the requested hand-painted Studio Ghibli aesthetic.
- − Colors are somewhat saturated, missing the soft pastel request.
FLUX.1 Kontext [max]
- + Excellent adherence to the 'hand-painted textures' and 'soft pastel' colors requested in the prompt.
- + Captures the soft, watercolor-like lighting typical of Studio Ghibli backgrounds.
- + Maintains the structural integrity of the source image while fully transforming the medium.
- − The man's facial expression is slightly more neutral/melancholic than the original exaggerated double-take.
Verdict: Both models successfully transformed the 'distracted boyfriend' meme while keeping the subjects recognizable. FLUX.1 Kontext [dev] creates a sharp, clean anime illustration, but FLUX.1 Kontext [max] far better interprets the specific stylistic requirements of the prompt by providing a soft, painterly texture and a nostalgic, dreamy color palette that genuinely evokes the Ghibli aesthetic.
Golden Hour Stroll
Image Editing“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”
AI Judge Analysis
FLUX.1 Kontext [dev]
- + Effectively alters the hair to show wind movement while preserving the subject's face.
- + Maintains high structural consistency with the original background and clothing.
- − The 'flying leaves' are very sparse and look more like small green specks than actual leaves.
- − The image feels less 'energetic' compared to the other model.
FLUX.1 Kontext [max]
- + Clearly adds a significant number of flying leaves throughout the scene to satisfy the prompt.
- + Adjusts the subject's posture and hand position to create a more 'lively' and active stride.
- + Successfully captures the hair blowing in the wind.
- − The changes to the subject's right hand (lower right on the dog) result in slightly unnatural finger anatomy.
- − Modifies the subject's posture more significantly than the source image.
Verdict: FLUX.1 Kontext [max] provided a much better interpretation of the 'energetic and lively' request by adding more leaves and slightly adjusting the subject's gait to imply movement. While FLUX.1 Kontext [dev] stayed truer to the original pose, its implementation of the flying leaves was too subtle to be effective.
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 Kontext [dev]
- + Clean vector emblem style with sharp lines.
- + Perfect text rendering including the accented character.
- + High contrast and centered composition.
- − Missed the 'banner' requirement for the date.
- − The cloche dome design looks slightly stylized like a building or crown rather than a standard kitchen cloche.
- − Lacks the requested 'subtle texture' on the background.
FLUX.1 Kontext [max]
- + Successfully included the 'banner' for the date as requested.
- + Effective use of subtle texture to achieve a vintage feel.
- + Sophisticated classic typography that matches the vintage theme better.
- − Minor artifact on the accent mark above the 'E' in Caffè.
- − The steam symbol is slightly simpler compared to the elaborate line work of the cloche.
Verdict: Both models performed well, but FLUX.1 Kontext [max] (Model B) followed the prompt more closely by including the specific 'banner' element and 'subtle texture' that FLUX.1 Kontext [dev] (Model A) missed. While Model A produced a cleaner vector look, Model B captured the aesthetic and design requirements of the vintage brief more comprehensively.
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 Kontext [dev]
- + Successfully adheres to the flat-vector iconography style requested.
- + Follows the NASA-inspired dark navy color palette well.
- + Captures the concept of a series of icon-based steps.
- − Significant text rendering failures, including a typo in the main header ('APOLO').
- − The iconography is messy and abstract, making the specific steps hard to identify.
- − Layout feels cluttered and lacks a clear logical flow.
FLUX.1 Kontext [max]
- + Excellent layout with a clear, logical progression through the mission steps.
- + High-quality vector illustrations of the Saturn V and Lunar Module.
- + Superior text legibility and accurate spelling of 'Apollo' and astronaut names.
- − Missed some of the specific requested steps in the sequence (e.g., descent).
- − Included terrestrial Earth textures that slightly deviate from the 'flat vector' request.
- − Layout becomes a bit repetitive with the two moon icons.
Verdict: FLUX.1 Kontext [max] is the clear winner as it produced a professional, legible, and aesthetically pleasing infographic with accurate spelling. In contrast, FLUX.1 Kontext [dev] struggled significantly with text rendering, even misspelling the primary mission name, and produced overly abstract icons that failed the informational purpose of the prompt.
Explore each model
Black Forest Labs' premium multimodal flow transformer with greatly improved prompt adherence and typography generation for in-context image generation and editing without compromise on speed