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
Settled by community votes across 17 shared challenges, with an AI judge weighing in on each.
FLUX.1 Kontext [max]
23.7
arena score
#23 of 62 in Text-to-Image
Skill signature
· Text-to-Image
Seedream 5.0 Lite
23.7
arena score
#22 of 62 in Text-to-Image
Vote tally
Where the votes landed
FLUX.1 Kontext [max]
0%
win rate
Ties
0%
Seedream 5.0 Lite
0%
win rate
Shared challenges
17
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 [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + Excellent handling of light refraction and caustic shadows on the table surface.
- + Highly detailed textures on the book and the sparkling blue sphere.
- + Photorealistic rendering of materials, especially the glass and wood grain.
- − The plant is quite dark and slightly less visible through the glass due to strong contrast.
Seedream 5.0 Lite
- + Clear visibility of the plant through the glass as requested in the prompt.
- + Simple, clean composition with soft, accurate lighting.
- + Correct spatial arrangement of all objects.
- − The glass lacks realistic thickness and refraction compared to Model A.
- − The textures are somewhat flat, particularly the blue sphere and the table.
Verdict: FLUX.1 Kontext [max] produces a much more professional and photorealistic image with complex lighting effects like caustics and detailed textures. Seedream 5.0 Lite follows the prompt accurately but lacks the visual depth and material realism found in the other model.
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 [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + Excellent depiction of rain with visible streaks and atmospheric depth
- + Highly realistic hand-skin texture and age spots
- + Strong adherence to the 'imperfect framing' prompt with a low, slightly off-center angle
- − The bicycle chain and pedal assembly is physically nonsensical
- − The subject's ethnicity is somewhat ambiguous compared to the specific prompt
- − Distracting background highlights that are a bit too busy
Seedream 5.0 Lite
- + Successfully captures the requested motion blur from passing cars
- + More accurate representation of an elderly Japanese man
- + Better mechanical coherence in the bicycle repair area
- − The rain effect is very subtle to the point of being nearly invisible in the foreground
- − The lighting on the subject feels slightly more artificial compared to the environment
- − The framing is quite standard and 'safe' despite the request for something imperfect
Verdict: FLUX.1 Kontext [max] captures the atmosphere of rain and the 'imperfect' cinematic framing much better, but fails significantly on the mechanics of the bicycle. Seedream 5.0 Lite provides a better character likeness and includes the requested motion blur, though it struggles to convincingly render the falling rain compared to its competitor.
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 [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + Exceptional skin texture with visible pores and realistic perspiration
- + High-quality intricate engraving on the plate armor
- + Dynamic lighting with strong warm reflections and atmospheric sparks
- − Missed the request for beads in the hair braids
- − The scars are very subtle and barely visible
Seedream 5.0 Lite
- + Accurately included small beads in the hair braids
- + Well-defined facial scars as requested in the prompt
- + Clean texture on the cloth underlayer and leather straps
- − The skin and hair textures appear somewhat synthetic or smoothed like a video game render
- − The sparks in the background are less integrated into the environment than Model A
Verdict: FLUX.1 Kontext [max] provides a much more lifelike and gritty image that captures the 'battle-worn' aesthetic through superior texture and lighting, though it missed the hair beads. Seedream 5.0 Lite followed more specific prompt details like the beads and visible scars, but the overall image quality feels more artificial and less cinematic. FLUX.1 Kontext [max] is the winner for its professional-grade visual quality and realistic materials.
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 [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + Excellent photo-realistic food quality
- + Elegant, sophisticated layout for a modern restaurant
- + Includes more representative menu items in a single grid
- − Garbled, unreadable text throughout the body
- − Doesn't clearly differentiate specific sections for appetizers and mains as requested
Seedream 5.0 Lite
- + Excellent text rendering with clear, legible English
- + Perfectly followed section instructions for appetizers, pizzas, and mains
- + Coherent graphic design with bold sans-serif fonts and vibrant accents
- − Food photos are slightly less detailed and more generic than Model A
- − Layout is quite a bit more basic and rigid
Verdict: FLUX.1 Kontext [max] produces much higher quality food imagery and a more authentic professional layout, but the text is completely illegible. Seedream 5.0 Lite followed every instructional requirement, including specific category headers and readable text, and is much closer to a functional menu design.
Magic Burger Explosion: Fiery Photorealism Challenge
Text-to-Image“Ad for 'Magic Burger'. Dynamic, exploded burger with all components (bun, patty, cheese, lettuce, tomato, sauce) suspended in mid-air. Emphasize photorealistic detail and a sense of motion. Dark, fiery background with glowing embers. Integrate text: 'MAGIC BURGER' as a prominent title, 'LIMITED TIME ONLY' as a secondary message, and '€6.99' in a starburst, all rendered with a fiery, glowing effect.”
FLUX.1 Kontext [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + Excellent typography with a theatrical glowing style
- + Highly detailed textures on the burger bun and patty
- + Good integration of embers and atmospheric effects
- − The burger is not truly 'exploded' as most components are still stacked
- − Failed to include the starburst for the price
- − Random bun halves floating on the sides feel disconnected
Seedream 5.0 Lite
- + Perfectly follows the 'exploded' instruction with vertically suspended components
- + Included all requested text elements, including the starburst
- + Dynamic composition with a strong sense of motion
- − Bottom bun has an unusual hole in the center resembling a bagel
- − Lighting on the burger is slightly less dramatic than the background
Verdict: Seedream 5.0 Lite followed the prompt much more accurately, successfully creating the 'exploded' view and including the requested starburst for the price. While FLUX.1 Kontext [max] had superior font rendering and texture detail, it failed the core compositional requirement of the prompt.
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 [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + Excellent text accuracy with no spelling errors.
- + Very realistic chalk texture with smudges and natural variations.
- + Perfect adherence to all requested menu items and prices.
- − The title is in print-style handwriting rather than the requested 'elegant cursive'.
Seedream 5.0 Lite
- + Features a more weathered, realistic chalkboard background.
- + Good chalk-like texture on the lettering.
- − Contains multiple spelling errors like 'Heriss', 'Beliter', and 'optoons'.
- − The text is written in a standard print style rather than the requested cursive for the title.
Verdict: FLUX.1 Kontext [max] is the clear winner as it followed the complex text prompt perfectly without any spelling mistakes. While both models failed to produce 'elegant cursive' for the title, Seedream 5.0 Lite suffered from poor text rendering accuracy, misspelling 'Herbs', 'Butter', and 'Options'.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
FLUX.1 Kontext [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + Excellent anatomical clarity and texture of the horse.
- + Very clean, cinematic lighting with realistic reflections on the visor.
- + Solid composition with a balanced background planet.
- − Failed the specific spatial instruction for the horse to be on top of the astronaut.
Seedream 5.0 Lite
- + Vibrant, colorful nebula background adds to the surreal quality.
- + High level of detail on the astronaut's suit electronics.
- − Failed the specific spatial instruction for the horse to be on top of the astronaut.
- − The horse's back legs and tail look slightly distorted and messy.
Verdict: Both FLUX.1 Kontext and Seedream 5.0 Lite failed the negative/compositional constraint 'horse on top, not vice versa', both producing a standard astronaut riding a horse. FLUX.1 Kontext is the superior image due to its cleaner rendering, more realistic textures, and better lighting, whereas Seedream 5.0 Lite has some structural issues with the horse's anatomy.
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 [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + High photographic realism with natural lighting and depth of field.
- + Excellent integration of the capybara's fur and clothing textures.
- + Captures the bored expression of the passenger well, despite being out of focus.
- − The perspective hides one of the paws on the steering wheel.
- − The passenger is blurred and partially obscured by the foreground framing.
Seedream 5.0 Lite
- + Directly follows the instruction for both front paws to be on the steering wheel.
- + The passenger is clearly visible looking at her phone as requested.
- + Wide composition captures the complete interior scene including the dashboard.
- − The capybara's head is disproportionately large compared to its body and the car interior.
- − The lighting on the capybara feels slightly artificial compared to the background.
- − Minor clipping artifacts around the hat and collar.
Verdict: FLUX.1 Kontext [max] produces a significantly more cinematic and photorealistic image with superior textures and lighting, making the scene feel more tangible. However, Seedream 5.0 Lite captures the specific details of the prompt better, clearly showing both the capybara's paws and the passenger's activity, despite its slightly awkward proportions and less realistic compositing.
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 [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + Perfectly rendered text with very few errors in typography.
- + Integrated layout that blends the central jack-o-lantern with the background successfully.
- + Highly detailed and consistent gothic border design.
- − Repetitive location text in the bottom section creating redundancy.
- − The scroll banner for the sub-text is less pronounced than requested.
Seedream 5.0 Lite
- + Excellent composition with a very clear scroll banner and parchment texture.
- + Strong adherence to the thorn and web border request.
- + Higher contrast lighting on the jack-o-lantern and background atmosphere.
- − Minor text clipping and overlapping in the top title.
- − The letter 'g' in 'frights' is slightly malformed on the scroll.
Verdict: FLUX.1 Kontext [max] provides superior text legibility and a more unified graphic design feel, though it stumbles with some redundant information at the footer. Seedream 5.0 Lite captures the requested artistic elements like the scroll and border with more visual punch but suffers from slight typographic artifacts.
Bald man challenge
Image EditingEdit instruction
“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”
FLUX.1 Kontext [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + Successfully added high density hair
- + Maintains the rugged outdoor lighting of the original
- − Significantly altered the facial features, making the eyes larger and the face narrower
- − Hair volume looks slightly unnatural and 'wig-like' at the top
Seedream 5.0 Lite
- + Excellent source preservation of the underlying facial geometry
- + Natural-looking hairline and realistic hair volume
- + Seamless integration with the existing beard
- − Slightly lowered the eye position compared to the original, though less than the alternative
Verdict: Seedream 5.0 Lite is the clear winner as it successfully adds a realistic head of hair while maintaining the core identity and fine details of the person in the source image. FLUX.1 Kontext [max] failed at source preservation, significantly altering the man's facial structure and eye shape, making it look like 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.”
FLUX.1 Kontext [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + Excellent text rendering with a stylized 3D effect that matches the scene
- + Superior shading and PBR textures on the fish and wood
- + Perfectly executed isometric perspective and lighting
- − Missed the small flag icon requested in the prompt
Seedream 5.0 Lite
- + Includes all requested elements including the flag icon
- + Offers a wider variety of sushi items within the scene
- + Very clean, minimal aesthetic that adheres well to the 'miniature' theme
- − Typography is less integrated into the 3D style, feeling more like a 2D overlay
- − The text and flag are not centered as requested, hugging the right side
Verdict: FLUX.1 Kontext [max] produced an image with superior visual quality and lighting, making the 3D 'miniature' effect feel more tangible, though it failed to include the flag icon. Seedream 5.0 Lite followed the text prompts more meticulously by including all elements, but it struggled with the layout and centering of the text assets. FLUX.1 is the winner for its professional rendering and artistic coherence.
Over-the-top cartoon caricature
EditingEdit instruction
“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.”
Source
FLUX.1 Kontext [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + Expertly captures the subject's outfit and hair color from the source image.
- + Includes all requested elements: news desk, microphone, dog, and hockey stick.
- + Strong caricature style with exaggerated features and expressive energy.
- − The hockey stick is partially cut off and awkwardly placed at the top edge.
- − Added large black glasses that were not in the source image.
Seedream 5.0 Lite
- + Excellent integration of the hockey stick and puck into the news anchor setting.
- + The caricature face is a very clean and recognizable translation of the source person.
- + The background and props create a more cohesive 'TV studio' environment.
- − The dog's paws and the hand holding it have slight anatomical glitches.
- − The facial expression is less 'exaggerated and humorous' compared to the more expressive Model A.
Verdict: Both models followed the instructions well, translating the subject into a stylized caricature while maintaining her denim shirt aesthetic. Seedream 5.0 Lite is the winner because it integrated the hockey theme more naturally into the composition and provided a cleaner, more complete visual of all requested elements, whereas FLUX.1 Kontext [max] had awkward cropping on the hockey stick.
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 [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + Excellent fur texture detailing on all four animals
- + Beautiful integration of butterflies within the lighting and depth of field
- + Accurate depiction of animal anatomy for all four requested species
- − The animals are mostly sitting still rather than 'chasing and tumbling'
- − The lighting includes artificial-looking starburst sparkles
Seedream 5.0 Lite
- + Better captures the 'chasing and tumbling' aspect of the prompt with more dynamic poses
- + Very clear god rays and dew sparkle effects on the grass
- + Good species representation for the kitten, puppy, bunny, and fox
- − The fox looks significantly more cartoonish/illustrated compared to the 'photorealistic' request
- − The kittens paws and the butterflies have slightly unrealistic edges
Verdict: FLUX.1 Kontext [max] provides a much higher level of photorealism and texture detail, making it look like a high-end photograph. Seedream 5.0 Lite captures the playful energy of the prompt better with tumbling poses, but it leans toward a 3D-animated aesthetic rather than the requested hyper-photorealistic style.
Studio Ghibli Anime Style
EditingEdit instruction
“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”
Source
FLUX.1 Kontext [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + Excellent hand-painted texture that feels like watercolor on paper.
- + Soft pastel color palette perfectly matches the requested mood.
- + Strong source preservation of the original poses and clothing patterns.
- − The facial features are slightly more generic anime than specific Studio Ghibli style.
- − The face of the woman in red has a slight misalignment in the eyes.
Seedream 5.0 Lite
- + Perfectly captures the specific Ghibli facial aesthetic and line art style.
- + Clean and vibrant colors while maintaining a nostalgic feel.
- + Exceptional preservation of the original image's composition and character expressions.
- − Lacks the requested 'hand-painted texture' found in Model A, appearing more digital.
- − The colors are a bit more saturated than the requested 'soft pastels'.
Verdict: Both models handled the meme transformation exceptionally well, but FLUX.1 Kontext [max] captured the requested hand-painted texture and soft pastel palette more accurately. Seedream 5.0 Lite, however, nailed the specific Studio Ghibli character design and facial features more convincingly, though it felt more like a modern digital animation than a traditional painting.
Golden Hour Stroll
Image EditingEdit instruction
“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”
FLUX.1 Kontext [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + Excellent character consistency with the original identity.
- + Natural-looking hair motion that flows with the wind.
- + Realistic leaf size and distribution scattered throughout the scene.
- − The leash loop significantly changed shape from the original.
- − Introduced an unnecessary hand modification where the left hand now touches the dog.
Seedream 5.0 Lite
- + Successfully added both blowing hair and flying leaves.
- + Preserved the original posture and leash shape well.
- − The flying leaves are unnaturally large and lack motion blur, appearing pasted on.
- − The hair edit looks slightly stiff and less integrated into the environment than the other model.
Verdict: FLUX.1 Kontext [max] provides a much more cohesive and lively edit with realistic wind-blown hair and leaves that fit the scale of the environment. While Seedream 5.0 Lite preserves the original composition more strictly, its added leaves are oversized and lack the dynamic feel requested in the prompt.
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 [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + Excellent typography with a hand-drawn, authentic vintage feel
- + Consistent woodcut-style texture across all elements
- + Strong composition with well-integrated banner and cloche
- − Minimalist request is interpreted more as a rustic illustration than a clean vector
Seedream 5.0 Lite
- + Reflects the minimalist and vector emblem style very well
- + Clean layout with legible text and simple iconography
- − The banner is not connected to the main icon, making it feel like floating elements
- − Typography is a bit generic compared to the vintage request
Verdict: FLUX.1 Kontext [max] creates a more cohesive and professional-looking logo with a beautiful hand-lettered aesthetic that perfectly fits the vintage theme. Seedream 5.0 Lite is more minimalist and clean, but the lack of integration between the banner and the cloche makes the design feel less finished.
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 [max]
Seedream 5.0 Lite
AI Judge Analysis
FLUX.1 Kontext [max]
- + Excellent typography and large, clear heading.
- + High-quality vector illustration style with a pleasant color palette.
- + Creative inclusion of crew silhouettes with accurate names.
- − Failed to follow the requested 6-step chronological sequence.
- − The layout of icons and labels is confusing and lacks logical flow.
- − Some graphic elements like the 'Lunar Orbit' icon are abstract and unrecognizable.
Seedream 5.0 Lite
- + Perfect adherence to the 6-step instructional sequence.
- + Very clean, consistent iconography that matches the flat-vector request.
- + Excellent layout that is easy to read and logically organized.
- − The 'Translunar' icon is a bit too minimalist compared to the others.
- − Small text for some crew names is slightly less legible than the main headings.
Verdict: While FLUX.1 Kontext [max] has more sophisticated artistic detail and beautiful crew silhouettes, it failed significantly on the instructional logic of the prompt. Seedream 5.0 Lite successfully followed all 6 specific steps in order with clear, consistent iconography, making it a much better infographic overall.
Next steps
Explore each model
ByteDance's image generation model with built-in reasoning, example-based editing, and deep domain knowledge, supporting up to 3K resolution