Black Forest Labs' 12-billion parameter multimodal flow transformer for in-context image generation and editing with character consistency, typography handling, and commercial-ready quality
Settled by community votes across 17 shared challenges, with an AI judge weighing in on each.
FLUX.1 Kontext [pro]
#41 of 62 in Text-to-Image
FLUX.2 [max]
#10 of 62 in Text-to-Image
Where the votes landed
FLUX.1 Kontext [pro]
0%
win rate
Ties
0%
FLUX.2 [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 [pro]
- + Excellent adherence to lighting instructions with a soft, natural window light.
- + The glass cube has highly realistic, thin edges typical of modern glassware.
- + The textures, especially the felt-like sphere and the wooden table, are very tactile.
- − The plant in the background is less distinct through the glass than requested.
- − The sphere appears to be floating rather than resting on the bottom surface.
FLUX.2 [max]
- + Features more realistic glass refraction and reflections, including the plant's visibility through the cube.
- + The red book has a convincing leather-like texture on the cover.
- + The use of a mirrored base inside the cube adds a visually interesting element of realism.
- − The lighting is a bit harsh with high-contrast shadows compared to the 'soft' light requested.
- − There are some minor digital artifacts/noise visible on the glass panels.
Verdict: Both models followed the complex spatial instructions perfectly. FLUX.1 Kontext [pro] creates a more natural, photographic atmosphere with soft lighting, while FLUX.2 [max] excels at the physics of glass, showing better refraction of the plant and more complex reflections. FLUX.2 [max] is the slight winner for its superior handling of the transparency and refractive properties of the glass cube.
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 [pro]
- + Excellent depiction of skin texture and rainy atmosphere
- + Strong shallow depth of field effect
- + The natural framing and motion blur on the street feel very candid
- − The man appears to be simply holding or mounting the bike rather than 'repairing' it
- − The rain texture looks a bit like a static filter overlay
FLUX.2 [max]
- + Perfect adherence to the 'repairing' part of the prompt with a kneeling pose and tools
- + Excellent motion blur on the background car
- + High level of detail on the bicycle and wet pavement reflections
- − The bike basket looks slightly distorted/misaligned
- − Visible AI artifact on the man's hands where he is touching the wheel spokes
Verdict: FLUX.2 [max] is the superior image as it captures the specific action of 'repairing' the bike with tools and a realistic kneeling posture, whereas FLUX.1 Kontext [pro] shows a man simply leaning on a bicycle. FLUX.2 [max] also better executes the motion blur and environmental storytelling of a rainy Japanese street.
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 [pro]
- + Excellent skin texture with realistic pores and subtle imperfections.
- + Natural-looking hair braids that flow logically from the scalp.
- + Warm, soft lighting that accurately reflects off the polished engraving.
- − The 'beads' requested in the hair are minimally represented by a single metal clasp.
- − The character looks less 'battle-worn' compared to the other model.
FLUX.2 [max]
- + Strong adherence to the 'battle-worn' prompt with significant dirt and scarring.
- + Includes many small beads woven into the braids as requested.
- + The armor engraving is very intricate and weathered.
- − The scar and dirt texture looks slightly painted on rather than integrated into the skin.
- − The lighting transition on the face is a bit harsh.
Verdict: Both models performed exceptionally well, but FLUX.2 [max] adhered more closely to the specific details of the prompt, including the beads in the hair and a clear 'battle-worn' aesthetic. FLUX.1 Kontext [pro] produced a more lifelike skin texture and natural lighting, but felt a bit too clean for the requested theme.
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 [pro]
- + Excellent typography with clean, bold sans-serif headers
- + Realistic food photography that blends well with the layout
- + Intuitive sectioning that matches the specific prompt categories
- − The 'pizza' text and images are misplaced, with a salad under the pizza header and a pizza under the mains header
- − Nonsense text for menu item descriptions
FLUX.2 [max]
- + Strict adherence to the 'grid' requirement for food photos
- + Clean professional layout with social media icons at the footer
- + Consistent pricing and sub-headers that look like a real menu
- − The image grid is repetitive and lacks variety in food types relative to the sections
- − The grid layout feels slightly cramped compared to model A
- − Minor text blurring in the smaller font sizes
Verdict: FLUX.1 Kontext [pro] creates a more visually striking and realistic design with superior typography, though it fails to align the food photos with their respective text headers. FLUX.2 [max] follows the 'grid' constraint much more literally and provides a more comprehensive menu structure, but the photos lack the unique character found in the first image. FLUX.1 Kontext [pro] is the likely winner due to its overall aesthetic quality and professional balance, which better suits the 'minimalist' request.
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 [pro]
- + Excellent typography with a vibrant neon glow effect
- + Very high photorealistic detail in the burger textures and melting cheese
- + Atmospheric background with high-quality embers and fire
- − Failed to include the starburst for the price
- − Repeated the price text twice, creating a cluttered bottom area
- − The burger is less 'exploded' and more just slightly separated
FLUX.2 [max]
- + Adhered perfectly to the request for an exploded view with all components suspended
- + Successfully included the starburst element for the price
- + Text styling perfectly matches the fiery prompt requirement
- − The lighting on the burger feels a bit flatter compared to the background
- − The sauce application looks slightly more artificial/digital than Model A
Verdict: While FLUX.1 Kontext [pro] offers superior photorealism in the food itself, FLUX.2 [max] is the winner because it followed the compositional instructions much better, correctly identifying the 'exploded' layout and the starburst element. FLUX.2 [max] also avoided the repetitive text error found in FLUX.1 Kontext [pro].
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 [pro]
- + Excellent chalk texture throughout the board
- + Accurately follows the item list including descriptive text at the bottom
- + High contrast and very clear legibility
- − The 'elegant cursive' requirement for the title is not fully met as it is mostly print blocks
- − Minor typo in the bottom text ('foiur' instead of 'four')
FLUX.2 [max]
- + Successfully uses elegant cursive handwriting for the text as requested
- + Excellent overall composition with atmospheric lighting and visible chalk dust smudges
- + Consistent and realistic handwriting style across all menu items
- − The pricing for the first item is missing a dash requested in the prompt style
- − Slightly less chalky texture in the strokes compared to Model A
Verdict: Both models followed the complex prompt with high accuracy, but FLUX.2 [max] is the winner for its superior interpretation of 'elegant cursive chalk handwriting' and its more atmospheric, realistic composition. While FLUX.1 Kontext [pro] had better chalk grit texture, it failed to provide the cursive style requested for the title and included a small typo.
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 [pro]
- + Perfectly follows the specific role-reversal prompt of horse-on-top.
- + Creative and surreal anatomical fusion, like the astronaut having hooves.
- + Cinematic lighting and clear, sharp resolution.
- − The small secondary astronaut on the horse's back adds confusion to the scale.
- − The satellite in the background is a bit blurry compared to the foreground.
FLUX.2 [max]
- + Beautiful composition with a vibrant galactic background.
- + Higher level of textural detail on the asteroid and spacesuit.
- + Stable, aesthetically pleasing layout.
- − Completely failed the negative constraint/specific instruction to put the horse on top.
- − Standard cliché interpretation rather than the requested surreal swap.
Verdict: FLUX.1 Kontext [pro] is the clear winner as it is the only model that followed the specific instruction to reverse the positions of the horse and astronaut. While FLUX.2 [max] produced a visually stunning image, it defaulted to the common trope of an astronaut riding a horse, ignoring the CORE requirement of the prompt.
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 [pro]
- + Near-perfect adherence to anatomy for capybara paws on the steering wheel.
- + High level of texture detail in the capybara's fur and the leather seating.
- + Tight framing creates a sense of being inside the vehicle.
- − The passenger is holding a phone but looking away, missing the 'looking at her phone' instruction.
- − The composition is a bit cramped, obscuring the taxi's interior details.
FLUX.2 [max]
- + Excellent composition that clearly shows the interior, the passenger, and the city lights through the window.
- + Perfect adherence to the passenger's behavior of looking at her phone with a bored expression.
- + High cinematic quality with realistic lighting shifts between the interior and exterior.
- − Serious anatomical error where the capybara has human hands instead of front paws.
- − The capybara's head looks somewhat pasted onto a human body rather than integrated.
Verdict: FLUX.1 Kontext [pro] creates a much more grounded and anatomically correct capybara, successfully rendering paws on the wheel as requested. However, FLUX.2 [max] captures the narrative of the prompt much better, specifically the bored businesswoman, though it fails significantly by giving the capybara human hands. FLUX.1 Kontext [pro] is the winner for avoiding the immersion-breaking 'uncanny valley' of human-handed animals.
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 [pro]
- + Excellent typography style that perfectly fits the 'vintage gothic' theme.
- + Captures the moody atmosphere well with the inclusion of a full moon and orange glow.
- + Detailed cobweb border work.
- − Includes hallucinated text: 'Your: Vorkleat: Iight & Spans'.
- − The date is rendered with a comma instead of a period (30.10,2026).
FLUX.2 [max]
- + Perfect adherence to all text fields with zero spelling errors.
- + Includes both thorns and webs in the border as requested.
- + The scroll banner is more clearly defined as a physical object.
- − The parchment texture is limited to the edges rather than the whole background.
- − The lighting on the pumpkin is a bit flat compared to the other model.
Verdict: While FLUX.1 Kontext [pro] captures a superior aesthetic and gothic font choice, it fails significantly by hallucinating extra messy text. FLUX.2 [max] is the preferred choice because it perfectly follows the text instructions for the event details while still maintaining a high-quality spooky composition with the requested thorns, webs, and scroll.
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 [pro]
- + Successfully added a thick, dense head of hair.
- + Matches the hair color well to the existing beard.
- + Maintains the overall lighting and environment.
- − Significantly altered the facial features and nose shape of the original person.
- − The hairline looks slightly artificial and too straight along the forehead.
FLUX.2 [max]
- + Excellent preservation of the original person's facial features and bone structure.
- + The hairstyle looks more natural with organic flow and texture.
- + Perfect integration of the hair with the existing ear and temple area.
- − No significant cons observed; the edit is highly successful and accurate.
Verdict: FLUX.2 [max] is the clear winner as it successfully adds the requested hair while perfectly preserving the identity of the person in the source image. In contrast, FLUX.1 Kontext [pro] creates a new face that only bears a passing resemblance to the original subject, failing the source preservation aspect of the task.
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 [pro]
- + Excellent typography style that matches the 3D aesthetic.
- + High-quality textures on the salmon and rice grains.
- + Warm, appealing lighting that enhances the food appearance.
- − The flag icon is stylized to the point of being a square with a circle, lacking traditional flag proportions.
- − The rice grains look slightly like bubbles or pearls rather than typical rice.
FLUX.2 [max]
- + Perfect adherence to the isometric perspective.
- + Includes a variety of sushi items which feels more complete for the prompt.
- + Superior typography layout and clear flag icon.
- − The text is plain 2D black font which clashes slightly with the 3D scene.
- − The lighting is somewhat flat compared to the other model.
Verdict: FLUX.2 [max] adhered better to the isometric layout and provided a more complete diorama with variety, including a clear flag icon. FLUX.1 Kontext [pro] produced a more visually striking single piece of sushi with better PBR textures, but its composition felt less like a miniature scene and more like a character render. FLUX.2 [max] is the winner for its superior layout and prompt adherence.
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 [pro]
- + Successfully converts the subject into a stylized caricature/comic art style.
- + Maintains the composition and clothing of the source image perfectly.
- − Completely ignores the specific requested elements (tv anchor job, dogs, and hockey).
- − The caricature is very basic and lacks the 'humorous' exaggeration requested.
FLUX.2 [max]
- + Includes every single requested element: tv anchor desk, dogs, and hockey sticks/arena.
- + Successfully creates a bobble-head style caricature that is humorous and exaggerated.
- + Preserves the subject's likness and clothing style while translating them into a new scene.
- − Text on the scoreboard and desk is nonsensical gibberish.
- − The source preservation is lower because it changed the background entirely, though this was necessary to follow the prompt.
Verdict: FLUX.1 Kontext [pro] failed the primary task, only providing a cartoon version of the face while ignoring the requested theme. FLUX.2 [max] perfectly followed the complex instructions, incorporating the hobbies and profession into a cohesive and funny caricature, even if it had to sacrifice the original background to do so.
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 [pro]
- + Warm and inviting color palette with strong god rays
- + Extremely soft and fluffy fur texture
- + Expressive and joyful facial expressions
- − Anatomical blending between the animals makes it hard to distinguish separate bodies
- − The bunny is rendered as a 'cat-bunny' hybrid with feline paws and face
- − Scale of the animals is inconsistent
FLUX.2 [max]
- + Successfully captures the active 'playfully chasing' and 'tumbling' motion from the prompt
- + Distinct and accurate anatomy for all four different animal species
- + Excellent inclusion of dew sparkles and lighting depth
- − Small minor artifact on the puppy's tail
- − The red fox's front paws have a slightly muddled silhouette against the grass
Verdict: While FLUX.1 Kontext [pro] has a beautiful color palette, it fails significantly on animal anatomy, creating strange hybrid creatures out of the bunny and kitten. FLUX.2 [max] follows the prompt much better, providing distinct species that are actually in motion and chasing butterflies as requested.
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 [pro]
- + Excellent preservation of the original composition and poses.
- + Captures the clean line work and specific character expressions of Ghibli-style anime well.
- + Vibrant but appropriate color palette that maintains the clear distinction between the characters.
- − The textures are a bit flat and lean more toward digital cell shading than a hand-painted look.
- − The background remains very sharp and literal compared to the requested 'dreamy' style.
FLUX.2 [max]
- + Successfully applies a soft, painterly texture that feels like a watercolor background.
- + Achieves a warm, nostalgic mood with soft lighting and a muted pastel color palette.
- + Excellent facial characterization that blends the original actors' features with a classic anime aesthetic.
- − The hand placement of the man is slightly more ambiguous and messy than the original.
- − Lost some of the sharp contrast that makes the 'distracted boyfriend' meme immediately recognizable.
Verdict: FLUX.1 Kontext [pro] does a better job of preserving the exact geometry and iconic poses of the meme, making it instantly recognizable. However, FLUX.2 [max] significantly better captures the 'Ghibli' atmosphere requested, using soft watercolor textures and a warm, nostalgic glow that feels much closer to the prompt's artistic requirements.
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 [pro]
- + Successfully added wind-blown hair and leaves.
- + Preserved nearly all details from the source image accurately.
- + Maintained consistent lighting and skin tones.
- − The wind effects look relatively static and subtle.
- − Minor distortion to the left hand grasping the leash loop.
FLUX.2 [max]
- + Stronger and more convincing wind-blown hair effect.
- + The addition of leaves is more plentiful and dynamic.
- + Stronger energetic feel that matches the prompt.
- − Slightly altered the woman's facial features compared to the source.
- − A small cluster of leaves overlaps her shoulder in a slightly unnatural way.
Verdict: Both models performed well at preserving the source image while adding motion, but FLUX.2 [max] is the winner for providing a more dynamic and visually impactful interpretation of the wind. While FLUX.1 Kontext [pro] was more faithful to the original face, the overall 'energetic feel' requested by the prompt was much better achieved by FLUX.2 [max].
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 [pro]
- + Excellent typography with a bold, professional feel.
- + Clean stamp-like texture that feels authentic to a vintage branding style.
- + Strong, clear contrast and minimalist vector composition.
- − Small typo in the banner ('EEST' instead of 'EST').
- − The cloche dome is very stylized and simple.
FLUX.2 [max]
- + Perfect text rendering with no spelling errors.
- + Elegant circular emblem composition that feels more cohesive.
- + Very subtle and pleasing paper texture in the background.
- − The line art is a bit thinner, lacking the 'punch' of a bold stamp logo.
- − The cloche dome illustration is slightly more generic.
Verdict: While both models executed the aesthetic well, FLUX.2 [max] is the winner because it successfully rendered all text without spelling errors, whereas FLUX.1 Kontext [pro] included an extra letter in 'Est.'. FLUX.2 [max] also provided a more complete 'emblem' feel through its circular frame and balanced layout.
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 [pro]
- + Features a visually engaging composition with a clear 'Apollo 11' title.
- + Uses the requested NASA-inspired color palette effectively.
- + Captures a sense of scale and movement within the trajectory lines.
- − Confuses logical labels, such as labeling the Earth as '4' and a Saturn-like planet as the start.
- − Text elements like 'Michael Collins' are misspelled or nonsensical ('2,far Collins').
- − Fails to follow the requested sequential 6-step infographic structure.
FLUX.2 [max]
- + Successfully implements a structured, step-by-step infographic layout as requested.
- + Renders legible and mostly accurate text for the crew names and mission stages.
- + Includes specific requested icons like the Lunar Module and the Earth with orbit rings.
- − Repetitive and incorrect labeling in the top headers (e.g., 'Earth Orbit' appears three times).
- − The composition feels a bit cramped within the vertical pillars.
- − The Saturn V rocket design is generic and less detailed than the one in the competing image.
Verdict: While FLUX.1 Kontext [pro] creates a more artistic and dynamic poster, it fails significantly on the instructional and logical aspects of the prompt, including planet identification and spelling. FLUX.2 [max] is the preferred choice because it understands the 'infographic' requirement, providing a clean, compartmentalized layout with legible text and icons that follow the requested sequence, despite some repetitive labeling errors.
Explore each model
Black Forest Labs' flagship image generation model delivering state-of-the-art quality with exceptional realism, precision, and consistency for both text-to-image and advanced image editing