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
Seedream 4.0
#15 of 62 in Text-to-Image
Where the votes landed
FLUX.1 Kontext [pro]
0%
win rate
Ties
0%
Seedream 4.0
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 photographic quality with realistic wood grain and soft lighting.
- + Perfectly captures the 'blue sphere inside the cube' and 'red book on top' spatial relationship.
- + Very clear and thin glass rendering which feels modern and elegant.
- − The plant is behind the cube but doesn't appear clearly visible through the glass panels themselves due to the angle.
- − The blue sphere has a slightly fuzzy, felt-like texture rather than a smooth finish.
Seedream 4.0
- + Excellent adherence to the 'plant visible through the glass' instruction.
- + Beautiful light rays and reflections on the wooden surface.
- + The blue sphere has a nice glass-like translucency that fits the scene's materials.
- − The cube geometry is slightly distorted, particularly on the left side where the edge doesn't align correctly with the base.
- − The red book looks slightly flat and lacks texture detail compared to the rest of the scene.
- − The perspective of the cube's base is slightly warped.
Verdict: FLUX.1 Kontext [pro] creates a more realistic and structurally sound image with superior textures and lighting, though it places the plant slightly further away. Seedream 4.0 follows the specific 'visible through the glass' instruction more literally but suffers from geometric distortion and less refined object details. FLUX.1 Kontext [pro] is the winner for its overall professional photographic quality and coherence.
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 natural skin texture and high-quality bokeh as requested.
- + Very realistic rendering of raindrops and wet clothing textures.
- + Captures the 'cinematic but realistic' atmosphere with professional lighting.
- − The man appears to be riding or leaning on the bike rather than actively 'repairing' it.
- − The framing is a bit more centered than 'imperfect framing' might suggest.
- − Failed to include the requested motion blur from passing cars.
Seedream 4.0
- + Adheres perfectly to the 'repairing' aspect of the prompt with tools and a crouched posture.
- + Successfully incorporates motion blur from a passing car in the background.
- + Generates excellent reflections on the wet pavement.
- − The resolution and facial details are significantly lower than Model A.
- − Visible AI artifacts, especially on the hands and the spokes of the wheels.
- − The lighting is somewhat flat and lacks the 'cinematic' quality requested.
Verdict: While FLUX.1 Kontext [pro] produced a stunning, high-fidelity portrait with superior textures, it missed the core action of 'repairing' and the specific detail of motion blur. Seedream 4.0 followed the prompt's instructions much more closely, including the repair tools, the posture, and the passing car, but suffered from poor image quality and structural artifacts. FLUX.1 Kontext [pro] is the winner for its professional aesthetics, despite the slight narrative drift.
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]
- + Extremely realistic skin texture and lifelike eye rendering
- + Sophisticated and elegant engraving on the plate armor
- + Subtle and professional lighting that highlights the metal without overexposure
- − Hair beads are very minimal and less noticeable than requested
- − Paladin looks more 'noble' than 'battle-worn' with very clean skin and hair
Seedream 4.0
- + Excellent adherence to the 'beads in hair' and 'battle-worn' (dirt/scars) descriptors
- + High-contrast torchlight creates a more dramatic atmosphere
- + Great detail on the leather straps and underlayer textures
- − Fire and bokeh effects are somewhat messy and distract from the character
- − Skin texture appears slightly more processed and 'digital' compared to Model A
Verdict: FLUX.1 Kontext [pro] produces a more photorealistic and technically polished portrait with superior skin and armor textures. However, Seedream 4.0 followed the specific prompt details much better, particularly the beads in the hair and the battle-worn aesthetic (dirt and scars) which are much more prominent.
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 adherence to the menu structure, including sections for appetizers, pizza, and mains.
- + Professional use of bold sans-serif fonts and typographic hierarchy.
- + Balanced composition that realistically mimics a printed menu layout.
- − Nonsense 'lorem ipsum' style text for descriptions and names.
- − Some visual inconsistency where a pizza image is placed under the 'Appetizers' section.
Seedream 4.0
- + High-quality, vibrant food photography.
- + Creative use of a grid layout for images.
- − Fails to create an actual menu, producing more of a mood board or collage.
- − Lacks specific menu text like item names, descriptions, or prices.
- − Poor layout for a functional menu, with text haphazardly placed around images.
Verdict: FLUX.1 Kontext [pro] successfully interpreted the prompt to create a functional restaurant menu with clear headers, prices, and a clean professional layout. Seedream 4.0 produced a visually appealing collage of food, but failed to include the necessary textual elements and structural organization requested for a 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.”
AI Judge Analysis
FLUX.1 Kontext [pro]
- + Excellent text rendering and typography for the primary title.
- + Accurate price rendering of €6.99.
- + High-quality, realistic textures on the patty and bun.
- − The burger is not really 'exploded' into individual components as requested.
- − Failed to include the starburst element for the price, repeating the price twice instead.
- − Composition feels static rather than dynamic.
Seedream 4.0
- + Successfully captured the 'exploded' dynamic view with separated components.
- + Includes the requested starburst element for the price.
- + Text is effectively integrated into the flames for a cohesive look.
- − Incorrect price rendering (€5.99 instead of €6.99).
- − The burger appear to have two separate bottom halves or stacks, creating a confusing structure.
- − The visual quality of the burger ingredients is less photorealistic and slightly more illustrative.
Verdict: FLUX.1 Kontext [pro] creates a much cleaner, more professional advertisement with perfect text, but fails to follow the layout instructions for an exploded view and starburst. Seedream 4.0 follows the creative composition requirements much better but fails on the specific data point (price) and has logical errors in the burger's construction. Seedream 4.0 is likely preferred for capturing the specific 'exploded' and 'starburst' aesthetic requested, despite the typo.
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]
- + Text is perfectly legible and clear.
- + Correctly followed the prompt for all three menu items including the price for the third item.
- + Consistent lettering style across the board.
- − The handwriting feels slightly too uniform and digital, lacking natural chalk smudging.
- − The title lacks the specific 'elegant cursive' style requested.
- − Contains significant spelling errors in the footer text ('ous gluten tree').
Seedream 4.0
- + Excellent chalk texture with realistic smudging and chalk dust.
- + Title features a much more distinct cursive handwriting style as requested.
- + Spelling in the footer is accurate and the composition feels more like a real cafe environment.
- − The text is slightly less crisp and 'rendered' than Model A's sharp lines.
- − The $9 price for the third item is slightly less clear due to the chalk texture overlay.
Verdict: Seedream 4.0 is the clear winner for its superior artistic interpretation of chalk texture and successful execution of the 'elegant cursive' title. While FLUX.1 Kontext [pro] provided very clear text, the overall aesthetic feels overly sterile and it failed significantly on the spelling of the footer text compared to Seedream 4.0's natural and accurate results.
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 depicts the complex spatial arrangement of the horse riding on top of the astronaut.
- + Included a secondary mini-astronaut on the horse for added surrealism.
- + High cinematic quality with realistic textures on the space suit and horse.
- − The astronaut's legs end in horse hooves, which was not requested.
- − The small astronaut rider's legs clipping through the horse's back.
Seedream 4.0
- + Dynamic composition with a powerful pose.
- + Good rendering of the nebula reflection in the helmet.
- − Failed the primary instruction to place the horse on top of the astronaut.
- − Anatomical issues with the horse's left front leg being detached/deformed.
- − The horse lacks a space suit or breathing apparatus in a vacuum environment.
Verdict: FLUX.1 Kontext [pro] followed the unconventional prompt instruction almost perfectly, displaying a horse physically positioned on top of an astronaut in space. Seedream 4.0 ignored the specific positioning request entirely, providing a standard 'astronaut riding a horse' image with significant anatomical errors in the horse's legs.
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]
- + Excellent photorealistic fur texture and lighting on the capybara.
- + Logical camera angle that makes the viewer feel like they are inside the car.
- + The woman is holding a phone as requested.
- − The woman in the back appears to be touching her face/hair rather than just looking at the phone.
- − The composition feels a bit cramped due to the extreme close-up.
- − Only one paw is clearly visible on the wheel.
Seedream 4.0
- + Successfully positions the woman in the back seat with a bored expression.
- + Higher overall image clarity and better background city bokeh.
- + Shows both paws on the steering wheel clearly.
- − The text on the cap says 'TAT' instead of 'TAXI'.
- − The capybara's paws look slightly human-like in texture.
Verdict: Seedream 4.0 is the winner as it follows the structural prompt instructions more accurately, showing the capybara with both paws on the wheel and the woman clearly in the back seat looking at her phone. While FLUX.1 Kontext [pro] has slightly more realistic fur rendering, its composition is too tight and the background character's action is less clear.
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 text legibility and alignment.
- + Sophisticated border design with fine spiderweb details.
- + Cinematic lighting on the central jack-o-lantern.
- − Includes a line of gibberish text ('Your: Vorkleat: Iight & Spans') not in the prompt.
- − The 'dark parchment' texture is less evident compared to the other model.
Seedream 4.0
- + Stronger adherence to the 'parchment poster' and 'thorns' prompt elements.
- + Better integration of the twisted trees and moody clouds into the composition.
- + Accurate text without adding extra nonsensical lines.
- − The text in the small scroll banner is slightly garbled and less legible.
- − Composition feels a bit more cluttered with the heavy thorn border.
Verdict: Both models followed the prompt well, but Seedream 4.0 followed the specific style instructions for thorns and parchment more accurately while avoiding the hallucinations of extra text found in FLUX.1 Kontext [pro]. FLUX.1 Kontext [pro] produced cleaner main typography and web details, but the inclusion of random text lines makes it less professional for a final design.
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]
- + Excellent hair texture and realistic styling
- + Perfect color match with the existing beard
- + Strong preservation of lighting and jacket details
- − Substantially altered the face shape and features (eyes and nose look different)
- − Background texture on the right has been slightly smoothed out
Seedream 4.0
- + Better preservation of original facial features
- + Excellent preservation of the background and surroundings
- + Adds hair that matches the request for a full head of hair
- − The hair texture looks soft and synthetic like a 1970s toupee
- − The hairline/forehead area looks smudged and lacks natural skin-to-hair transition
Verdict: While FLUX.1 Kontext [pro] produced much more realistic hair texture, it failed the preservation criteria by altering the person's face significantly. Seedream 4.0 successfully kept the person's original face and the background intact, though the added hair lacks the fine detail and natural hairline found in the other model.
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 with a charming 3D cartoon style
- + Perfectly followed the 'small raised diorama' instruction with a clean wooden block
- + Superior lighting and material textures that match the 'soft refined' request
- − The rice grains are overly rounded and pebble-like, losing some realism
Seedream 4.0
- + Includes a wider variety of sushi items
- + Excellent isometric perspective and centering
- + Accurate Japanese flag icon
- − The font is basic and less integrated into the 3D scene compared to Model A
- − The lighting is a bit harsh, creating deep shadows that contrast with the 'gentle lighting' request
Verdict: FLUX.1 Kontext [pro] produced a more cohesive 3D aesthetic with high-quality stylized typography that feels like a professional miniature. While Seedream 4.0 provided more variety in the sushi itself, its layout and basic text rendering were less polished than FLUX.1 Kontext [pro]'s refined diorama approach.
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 creates high-quality caricature artwork based on the subject's features.
- + Maintains the outfit and basic selfie composition of the source image.
- − Completely fails to include the requested profession (TV anchor), hockey theme, or dogs.
- − The background is replaced with a generic solid color rather than preserved.
Seedream 4.0
- + Successfully incorporates all requested elements: TV anchor desk/mic, a dog, and hockey gear.
- + Preserves the overall room layout and furniture from the source image remarkably well.
- + Uses a classic caricature style with an exaggerated head and facial expression.
- − There are minor artifacts on the hands holding the phone and the microphone.
- − The hockey stick is positioned at a strange angle relative to the desk.
Verdict: FLUX.1 Kontext [pro] failed the prompt instructions by ignoring all the thematic requests, delivering only a stylized portrait. Seedream 4.0 followed all instructions perfectly, creating a humorous caricature that integrated the TV anchor, dog, and hockey elements while maintaining a high level of fidelity to the original background.
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]
- + Excellent fur texture rendering and soft lighting contrast
- + High aesthetic appeal with a clean, professional depth of field
- − Failed to include a rabbit, instead showing two kitten-like creatures with slight ear variations
- − Includes 4 animals but they are sitting still rather than 'playfully chasing and tumbling'
Seedream 4.0
- + Perfect prompt adherence, including the golden retriever, tabby kitten, bunny, and red fox kit
- + Captures the action of 'tumbling and chasing' much better with dynamic poses
- + Beautiful representation of dew sparkles and god rays
- − The fox's anatomy is slightly awkward as it tumbles
- − Small artifacts in the fur along the edges of the moving animals
Verdict: While FLUX.1 Kontext [pro] produced a more polished and visually clean image, it failed to correctly generate the specific animals requested, omitting the bunny. Seedream 4.0 followed the prompt much more accurately, successfully including all four specific species and capturing the active, playful energy of the 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 [pro]
- + Excellent preservation of the original subjects' poses and facial expressions.
- + Strong watercolor texture and linework that authentically mimics Ghibli's style.
- + Accurately replicates the background architecture and clothing patterns from the source.
- − The colors are slightly more saturated than the requested 'soft pastel' palette.
Seedream 4.0
- + Perfectly captures the requested soft pastel color palette and dreamy lighting.
- + Achieves a very convincing hand-painted watercolor aesthetic.
- + Preserves the core composition and character recognizable from the meme.
- − Lost the background architectural details, replacing them with a vague sky.
- − The facial expressions are significantly muted compared to the intense emotions in the original image.
Verdict: FLUX.1 Kontext [pro] is the winner for its superior ability to preserve the specific details and emotional intensity of the original 'distracted boyfriend' meme while applying a high-quality Ghibli-style filter. While Seedream 4.0 captures the 'soft pastel' color request better, it simplifies the background too much and loses the nuances of the characters' expressions.
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]
- + Excellent preservation of the subject's facial features and the dog's characteristics.
- + Successfully adds wind-blown hair that looks natural and integrated into the scene.
- + Maintains high sharpness and clarity throughout the image.
Seedream 4.0
- + Stronger emphasis on the requested 'energetic' feel with more numerous falling leaves.
- + Adds a lens flare/lighting effect that contributes to a lively atmosphere.
- − Noticeable degradation in texture quality, resulting in a slightly blurry or 'soft' look compared to the original.
- − The dog's face is slightly altered, losing some of the sharp detail from the source image.
- − The hair edit looks more artificial and stringy compared to Model A.
Verdict: FLUX.1 Kontext [pro] is the winner as it successfully applied the motion edits while maintaining the high resolution and specific details of the source image. Seedream 4.0 added more leaves and lighting effects but suffered from a significant loss in clarity and slightly altered the facial features of both the human and the dog.
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 rendering with the requested accent mark.
- + Strong minimalist vector aesthetic that feels contemporary yet vintage.
- + Very clean application of the grainy texture on a light background.
- − Includes a spelling error in the banner text ('EEST' instead of 'EST').
- − The steam icon is very simplified, appearing more like a sprout.
Seedream 4.0
- + Accurate spelling of all requested text including 'Est. 1720'.
- + Beautifully rendered cloche icon with realistic shading and elegant steam.
- + Excellent composition with the arched typography framing the central element.
- − The cream background texture creates slight artifacts/blurring around some letters.
- − Slightly less 'minimalist' than requested due to the detailed shading on the cloche.
Verdict: Seedream 4.0 is the clear winner as it correctly followed the text requirements, whereas FLUX.1 Kontext [pro] included a significant typo ('EEST'). Seedream 4.0 also provided a more aesthetically pleasing cloche and better overall balance in the emblem's composition.
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]
- + Excellent typography and artistic vintage NASA feel.
- + Good use of 'muted red' and navy palette.
- + More natural integration of the lunar surface in the composition.
- − Confused astronomical logic, depicting Saturn-like rings around a generic planet and the Moon.
- − Failed to follow the requested numbered step-by-step sequence.
- − Text and labels are disorganized and contain typos like '2, f ar Collins'.
Seedream 4.0
- + Successfully followed the 6-step infographic structure with numbered labels.
- + Included specific astronaut profiles as requested by the spirit of the prompt.
- + Accurately rendered icons for Earth, the Moon, and the Lunar Module.
- − Visible typo 'Descent Surfcee' in the fifth step.
- − Background stars are a bit large and distract from the 'clean' vector aesthetic.
- − Text 'Landing 6' is oddly balanced and overlaps the layout slightly.
Verdict: Seedream 4.0 is the clear winner for its adherence to the infographic structure, accurately representing all six requested steps in a logical sequence. While FLUX.1 Kontext [pro] has a more sophisticated artistic style, it completely failed the logic of the mission by adding Saturn rings to lunar orbits and failing to format the steps as a readable guide.
Explore each model
ByteDance's image generation model with integrated text-to-image and image editing capabilities in a unified architecture, supporting up to 4K resolution