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.2 [max]
#10 of 62 in Text-to-Image
Where the votes landed
FLUX.1 Kontext [dev]
0.0%
win rate
Ties
0.0%
FLUX.2 [max]
100.0%
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 adherence to lighting instructions with a clear window light source on the left.
- + Vivid colors and sharp focus on the central subject.
- + Realistic reflection of the sphere on the base of the cube.
- − The glass cube is missing its front face, appearing more like a glass stand or table.
- − The plant is very close and dominant, reducing the focus on the specified 'partially visible' effect.
FLUX.2 [max]
- + Perfect structural representation of a glass cube with all edges clearly visible.
- + High-quality leather texture on the red book.
- + Good spatial arrangement that keeps the plant behind the cube as requested.
- − The blue sphere appears slightly translucent or has odd internal reflections that make it look less 'solid'.
- − Includes ghost-like lens flares or extra reflective spheres appearing outside the cube that weren't requested.
Verdict: FLUX.2 [max] is the preferred image because it correctly renders the glass as a complete cube, whereas FLUX.1 Kontext [dev] produced an object that looks like an open-sided stand. While both models followed the spatial and color instructions well, FLUX.2 [max] shows superior understanding of 3D geometry and material textures.
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]
- + Excellent shallow depth of field effect with realistic bokeh
- + Clean, high-resolution rendering of the subject's face
- − Subject is posing with the bike rather than repairing it
- − Composition feels centered and intentional, missing the 'imperfect framing' request
- − The rain effect looks like a simple overlay rather than interacting with the environment
FLUX.2 [max]
- + Accurately depicts the 'repairing' action with tools on the ground
- + Highly realistic skin texture on the hands and face
- + Effective motion blur on the passing car and authentic 'imperfect' street photography framing
- − The handlebar structure is slightly anatomically incorrect for a bicycle
- − The bicycle basket has some minor wire-rendering artifacts
Verdict: FLUX.2 [max] significantly outperformed FLUX.1 Kontext [dev] by adhering to almost every nuanced part of the prompt, specifically the 'repairing' action and the 'imperfect framing.' While FLUX.1 Kontext produced a clean image, it failed the core narrative of the prompt, showing a man simply standing with a bike.
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 warm lighting with strong golden-hour rim lights on the hair
- + Intricate high-relief engravings on the breastplate
- + Balanced composition that emphasizes the character's gaze
- − Missed the prompt instruction for braided hair with beads (hair is mostly slicked back)
- − Skin is too clean for a 'battle-worn' description with minimal dirt/scars
FLUX.2 [max]
- + Perfect adherence to specific details like braided hair with beads
- + Realistic skin texture with multiple scars and heavy dirt
- + Exceptional material definition on the weathered leather straps and fraying cloth layer
- − The lighting is a bit cooler and more diffuse than the 'warm torchlight' requested
- − The background bokeh sparks are slightly less defined than in Image A
Verdict: FLUX.2 [max] significantly outperformed the other model in prompt adherence, strictly following instructions for braided hair, beads, and battle-worn skin details that FLUX.1 Kontext [dev] ignored. While FLUX.1 Kontext [dev] produced a more cinematically lit image with higher contrast, FLUX.2 [max] captured the specific 'gritty' texture and complex accessory details much more accurately.
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 photography of food
- + Strong, bold typography with a modern aesthetic
- − Poor information architecture that doesn't resemble a functional menu
- − Text is largely gibberish and poorly distributed
FLUX.2 [max]
- + Excellent adherence to the menu structure with clear sections
- + Very clean, professional, and practical minimalist layout
- + Contains pricing and descriptions that mimic a real product
- − Lower visual intensity in the food photography compared to the other model
- − Small icons and footer text are slightly warped
Verdict: FLUX.2 [max] significantly outperformed in following the prompt's structural requirements, creating a logical menu with clear sections for pizza and mains. While FLUX.1 Kontext [dev] had more vibrant photography, its layout failed to function as a usable menu design, prioritizing a collage-style look over the requested professional document layout.
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 for the main title
- + High-quality, vibrant lighting on the coals and background
- − Failed to create an 'exploded' burger, showing a mostly assembled one instead
- − Text error in the secondary message: 'LNHLY' instead of 'ONLY'
FLUX.2 [max]
- + Successfully followed the 'exploded' instruction with components suspended in mid-air
- + Perfect text accuracy for all required messages
- + Superior adherence to the specific 'fiery, glowing effect' on the text
- − The starburst graphic for the price is slightly less polished than the rest of the image
Verdict: FLUX.2 [max] is the clear winner because it accurately followed the core structural request for an 'exploded' burger with suspended components, whereas FLUX.1 Kontext [dev] generated a standard, intact burger. Additionally, FLUX.2 [max] maintained perfect text spelling and better captured the requested fiery aesthetic for the typography.
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]
- + Successfully captured a variety of handwritten styles on a chalkboard.
- + Good centering and layout of the text elements.
- − Numerous spelling errors including 'Mashroom', 'Risoktso', and 'Octpus'.
- − Failed to render the date correctly, resulting in garbled text.
- − Included repetitive words like 'with with' and messy price overlaps.
FLUX.2 [max]
- + Excellent prompt adherence with nearly perfect spelling of complex menu items.
- + Superior chalk texture and realistic handwriting variation across the board.
- + Properly rendered the requested date and cursive title style.
- + Atmospheric lighting and smears add to the realism of a cafe setting.
- − The cursive for 'Risotto' and 'Octopus' is slightly condensed, though still legible.
Verdict: FLUX.2 [max] is the clear winner as it followed every instruction, including specific dates and complex spellings, with high realism. FLUX.1 Kontext [dev] struggled significantly with the text, producing several typos and nonsensical characters in place of the date.
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 and the astronaut to be the mount
- + High anatomical clarity in the face and hands of the astronaut
- + Clean, cinematic lighting that highlights the textures of the space suit
- − The horse's hind legs appear somewhat merged with the astronaut's back
- − The background is quite empty compared to the cinematic prompt request
FLUX.2 [max]
- + Beautiful cosmic background with a detailed galaxy and asteroid
- + Excellent texture on the horse and space suit
- + High sense of realism and scale
- − Failed the negative constraint/positional instruction; the astronaut is riding the horse
- − The astronaut's left leg has anatomical clipping issues with the horse's flank
Verdict: FLUX.1 Kontext [dev] is the clear winner because it successfully interpreted the difficult conceptual prompt of a 'horse riding an astronaut,' whereas FLUX.2 [max] ignored the specific positional instruction and generated a standard astronaut riding a horse. While FLUX.2 [max] has a more vibrant background, FLUX.1 Kontext [dev] demonstrated much higher prompt adherence and logical understanding of the surreal request.
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]
- + Excellent texture on the capybara's fur
- + The passenger has a perfect bored expression as requested
- + Strong photorealistic quality and lighting
- − The capybara only has one paw on the steering wheel instead of both front paws
- − The capybara's hands look more like primate/human hybrids with long claws
FLUX.2 [max]
- + Successfully placed both 'paws' on the steering wheel
- + Excellent composition showing more of the taxi interior and city lights
- + The hat is more distinctively a traditional taxi driver cap
- − The hands on the steering wheel are human hands rather than capybara paws
- − The passenger lacks the requested bored expression and looks more focused on the screen
Verdict: Both models struggled with the anatomical challenge of a capybara's paws on a steering wheel, with FLUX.1 Kontext [dev] creating hybrid claw-hands and FLUX.2 [max] giving the animal human hands. FLUX.1 Kontext [dev] captured the desired atmosphere and passenger expression much better, though FLUX.2 [max] had a superior composition of the taxi's interior.
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]
- + Features a consistent thorn border around the entire image.
- + Large, bold heading that is easy to read.
- − Several spelling errors in the scroll banner and location text.
- − Lacks the 'dark parchment' and 'moody night sky' textures requested, looking more like a graphic vector.
- − Jack-o-lantern and bats are very simplistic and cartoonish.
FLUX.2 [max]
- + Excellent text rendering with no spelling mistakes and appropriate gothic fonts.
- + High-quality 'cinematic' lighting and atmospheric background with trees and mist.
- + Successfully incorporates all requested elements including the parchment texture and scroll.
- − The thorn border is a bit messy and overlaps with the spider webs in the corners.
Verdict: FLUX.2 [max] is the clear winner as it followed every instruction, including the difficult text requirements, while maintaining a high-quality cinematic aesthetic. FLUX.1 Kontext [dev] struggled with spelling and produced a much coarser, more cartoonish image that lacked the atmospheric depth requested.
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 adds a full head of hair with thick volume.
- + Maintains the general color palette and composition of the original.
- − Reduces the man's age significantly, making him look several decades younger.
- − Changes core facial features including the eyes, nose shape, and skin texture.
- − Changes the glasses frames and smooths out the rugged details of the face.
FLUX.2 [max]
- + Excellent source preservation, keeping the original man's face, wrinkles, and age intact.
- + Adds hair that matches the texture and messy aesthetic of the existing beard.
- + Maintains the exact glasses and lighting from the source image.
- − The transition from the hair to the forehead (hairline) is slightly messy and feathered.
Verdict: FLUX.2 [max] is the superior choice for this editing task because it successfully adds the requested hair while preserving the identity and age of the subject. FLUX.1 Kontext [dev] fails as an edit model by completely changing the man's face and de-aging him, resulting in a different person entirely.
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]
- + Clean, bold text rendering with excellent legibility.
- + Soft, clay-like textures match the cartoon aesthetic well.
- + Very high clarity and centered composition.
- − The flag icon is abstract and does not represent the Japanese flag.
- − The sushi anatomy is slightly nonsensical, with rice appearing as large white lumps inside a black band.
- − The perspective is more front-on than the requested 45-degree top-down isometric view.
FLUX.2 [max]
- + Perfectly executes the 45-degree isometric perspective on a tiered diorama base.
- + Includes a correct Japanese flag icon as requested.
- + High level of detail in the sushi variety, wasabi, and ginger while maintaining a clean aesthetic.
- − Text is slightly less 'bold' compared to Model A, though still very clear.
- − Shadows are a bit sharper, losing some of the 'gentle' lighting feel requested.
Verdict: FLUX.2 [max] is the clear winner as it perfectly adheres to the technical framing requirements like the 45-degree isometric view and the tiered diorama base. While FLUX.1 Kontext [dev] produced a high-quality graphic, it failed on the specific flag icon and the isometric perspective, whereas FLUX.2 [max] delivered a more comprehensive interpretation of the prompt's details.
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 captures a hand-drawn caricature art style
- + Preserves the subject's denim clothing and basic facial structure well
- + Includes the dog and TV elements requested in the prompt
- − Completely misses the 'hockey' requirement of the prompt
- − Text rendering is poor and nonsensical
- − Visual quality is a bit flat compared to Model B
FLUX.2 [max]
- + Excellent adherence to all prompt elements including TV anchor desk, dogs, and hockey sticks/arena
- + High visual quality with a polished, professional caricature aesthetic
- + Creative use of composition to blend the ice rink with a newsroom
- − Changes the subject's eye color from brown to blue
- − Source preservation is lower as it changes the pose and camera angle significantly
Verdict: FLUX.2 [max] is the clear winner as it successfully incorporated all three required elements (TV anchor, dogs, and hockey) into a cohesive and visually appealing caricature scene. FLUX.1 Kontext [dev] produced a simpler drawing that failed to include the hockey theme and had significant issues with garbled text.
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]
- + Excellent vibrant colors and joyful expressions
- + Good focus on the three main subjects
- + Clean, high-contrast lighting
- − Failed to include all requested animals (missing the rabbit and fox kit)
- − Anatomy issues with the middle kitten's legs
- − The background is very blurry and lacks the 'lush wildflower' variety requested
FLUX.2 [max]
- + Perfect adherence to all requested animal types (puppy, kitten, rabbit, fox)
- + Stunning lighting effects with god rays and dew sparkles as requested
- + Detailed environmental rendering with a rich variety of wildflowers
- − The fox kit has slightly unnatural human-like pupils
- − Minor blending issues where the rabbit's paws meet the grass
Verdict: FLUX.2 [max] is the clear winner as it successfully rendered all four specific animals mentioned in the prompt, whereas FLUX.1 Kontext [dev] omitted two of them. Furthermore, FLUX.2 [max] better captured the environmental details like the dew sparkles and variety of wildflowers, creating a much more complete and high-quality scene.
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]
- + Excellent preservation of the source image's composition and poses
- + Clean line art and vibrant colors
- − Style leans more towards generic modern anime rather than Studio Ghibli
- − Lighting is flat and lacks the 'dreamy' quality requested
FLUX.2 [max]
- + Perfectly captures the Studio Ghibli aesthetic with soft watercolor textures
- + Excellent use of pastel colors and warm, nostalgic lighting
- + Successfully translates the facial expressions into a more whimsical style while keeping the meme recognizable
- − Slight loss of sharpness in the background compared to the foreground characters
Verdict: While both models successfully transformed the iconic meme into an illustration, FLUX.2 [max] far exceeded the prompt requirements by providing the specific painterly, nostalgic texture associated with Studio Ghibli. FLUX.1 Kontext [dev] produced a clean anime-style image, but it lacks the 'hand-painted' texture and 'dreamy' lighting requested in the instructions.
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]
- + Excellent preservation of the source image facial features and overall landscape.
- + Realistic subtle hair movement that feels integrated with the scene.
- − The 'flying leaves' are barely noticeable and look like small green specks.
- − The overall sense of energy and motion is very low compared to the prompt requirements.
FLUX.2 [max]
- + Successfully creates a high-energy, dynamic feel with dramatic hair motion.
- + Abundant falling leaves in the foreground and background fulfill the prompt effectively.
- + Preserves the identity of the person and dog remarkably well despite the significant additions.
- − Some leaves in the extreme foreground are a bit blurry, though this adds to the depth of field.
- − Slight modification to the dog's expression/mouth compared to the source image.
Verdict: FLUX.2 [max] is the clear winner as it much more effectively interprets the 'dynamic' and 'lively' aspects of the prompt, providing dramatic hair movement and clearly visible flying leaves. FLUX.1 Kontext [dev] was too conservative with its edits, resulting in a static image that barely suggests motion.
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]
- + Excellent typography rendering with perfect accuentuation
- + Clean and bold minimalist vector style
- + High contrast and strong visual presence
- − Failed to include the requested banner for the 'Est. 1720' text
- − Cloche illustration is slightly abstract and looks a bit like a dome or a cupcake liner
- − Lacks the requested 'subtle texture'
FLUX.2 [max]
- + Successfully followed all instructions including the banner and steam
- + Beautiful subtle paper texture on the background
- + Sophisticated emblem composition with arched typography
- − The 'E' in 'CAFFÈ' is slightly thinner/weaker than official classic typography standards
- − Line weight for the steam is a bit thin compared to the rest of the illustration
Verdict: FLUX.2 [max] is the winner as it perfectly adhered to the complex prompt requirements, including specific elements like the banner and the subtle texture while maintaining a professional vintage aesthetic. FLUX.1 Kontext [dev] produced a clean logo, but it missed the banner element and favored a more modern, flat design over the requested vintage emblem style.
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]
- + Captures the navy and muted red NASA-inspired color palette well
- + Icons have a cohesive, minimalist vector aesthetic
- − Significant spelling errors in title and descriptive text
- − The sequence of mission steps is disjointed and non-linear
- − Text becomes unreadable gibberish toward the bottom
FLUX.2 [max]
- + Excellent layout that logically follows the requested mission steps from left to right
- + Clean, high-quality typography with accurate spelling of astronaut names
- + Iconography effectively matches the prompt requirements for Saturn V, Moon, and Lunar Module
- − Repeats Earth Orbit label incorrectly across several panels
- − Visual style is slightly more detailed than a strict flat-vector style
Verdict: FLUX.2 [max] significantly outperforms FLUX.1 Kontext [dev] by providing a coherent infographic structure that follows the requested mission steps. While FLUX.2 [max] has some repetitive labeling issues, FLUX.1 Kontext [dev] suffers from major spelling errors and fails to organize the information in a logical or useful way.
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