Head to head
Esc

Models · slot A

to navigate to pick

FLUX.2 [dev] Black Forest Labs Seedream 4.0 ByteDance

Settled by community votes across 20 shared challenges, with an AI judge weighing in on each.

FLUX.2 [dev]

24.3 arena score

#19 of 62 in Text-to-Image

Skill signature · Text-to-Image

Seedream 4.0

24.7 arena score

#15 of 62 in Text-to-Image

Vote tally

Where the votes landed

FLUX.2 [dev]

25.0%

win rate

Ties

0.0%

Seedream 4.0

75.0%

win rate

25.0% 0.0% ties 75.0%
Shared challenges 20

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.2 [dev]
Seedream 4.0
0% wins 0% ties 100% wins

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent handling of glass thickness and optical refraction.
  • + The blue sphere is correctly placed inside the volume of the cube.
  • + Soft lighting feels natural and consistent across the table and glass surfaces.
  • The sphere appears to be floating slightly rather than sitting on the bottom.
  • The plant in the background is very blurry, making it harder to see through the glass as requested.

Seedream 4.0

  • + The cube geometry is very sharp and precise.
  • + Reflections on the mirrored base of the cube add a nice level of detail.
  • + The plant is clearly visible through the glass panels.
  • The red book is not perfectly centered and looks a bit small for the cube.
  • The perspective of the table edges is slightly inconsistent with the cube's orientation.

Verdict: Both models followed the complex spatial instructions perfectly. FLUX.2 [dev] produced a more photorealistic image with superior light and material physics, while Seedream 4.0 provided a cleaner, more literal interpretation of the 'visible through the glass' instruction. FLUX.2 [dev] is the winner for its more convincing textures and natural integration of the objects.

Man and Car in California

Editing
Edit instruction

“Make a photo of the man driving the car down the California coastline”

Source
FLUX.2 [dev]
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent preservation of the subject's specific outfit including the patterned coat and knit scarf.
  • + High realism in interior car textures and cinematic lighting.
  • + The hairstyle is very close to the source image.
  • The car interior is somewhat generic and loses the specific dashboard look of the source Rolls Royce.
  • The man's facial features have changed significantly from the source.

Seedream 4.0

  • + Successfully merges the car and man while preserving most of the car's exterior design.
  • + Good overall composition that clearly shows the coastline and the car in motion.
  • + Maintains the man's likeness and hair structure reasonably well.
  • The man is disproportionately small relative to the car interior.
  • Noticeable digital artifacts and 'ghosting' around the man's head and the coastline.
  • Clothing details like the coat pattern are blurred and less distinct than in the source.

Verdict: FLUX.2 [dev] produces a much higher quality, more realistic image with superior detail preservation in the clothing and lighting, though it changes the man's face. Seedream 4.0 does a better job of keeping the specific car's exterior and the man's likeness, but suffers from poor scaling of the subject and significant visual artifacts.

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.2 [dev]
Seedream 4.0
0% wins 0% ties 100% wins

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent depiction of motion blur with multiple passing cars that look realistic
  • + Highly detailed skin textures and lifelike facial features
  • + Very authentic rain effects and complex reflections on the asphalt
  • The bicycle frame geometry is slightly anatomically incorrect near the handlebars
  • The man's right leg appears to be missing or merged into the bike frame

Seedream 4.0

  • + Captures a more traditional Japanese street vibe with tool details on the ground
  • + The bicycle structure is more coherent and recognizable
  • + Good use of puddle reflections and composition
  • Skin and hair textures look slightly smoothed and less like a 'natural' 50mm photo
  • The motion blur on the car is a bit muddy and looks like a digital filter rather than long exposure
  • Rain is barely visible compared to the request for light rain

Verdict: FLUX.2 [dev] significantly outperforms Seedream 4.0 in terms of photorealism, particularly in the rendering of the man's skin and the sophisticated motion blur of the traffic. While Seedream 4.0 has a more structurally sound bicycle, FLUX.2 [dev] captures the cinematic, high-fidelity aesthetic and the specific 'light rain' atmosphere requested in the prompt more effectively.

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.2 [dev]
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent depiction of ornate, engraved plate armor with clear floral motifs.
  • + Detailed beads in the braids that match the prompt perfectly.
  • + Authentic and balanced warm lighting from the visible torch.
  • The scars look slightly like surface level paint rather than healed tissue.
  • The straps are a bit clean for a battle-worn character.

Seedream 4.0

  • + Superior gritty skin texture with more realistic dirt and faint scarring.
  • + Dynamic bokeh sparks and dramatic hair lighting.
  • + Better portrayal of the 'battle-worn' aesthetic on the armor's surface.
  • The torch in the background has a slightly messy, blown-out digital glow.
  • Engravings on the armor are less distinct and intricate than Model A.
  • Some minor artifacts in the hair/bead transitions.

Verdict: Both models followed the prompt exceptionally well, capturing the paladin aesthetic with high detail. FLUX.2 [dev] stands out with its incredibly crisp, high-contrast armor engravings and well-defined braids, whereas Seedream 4.0 excels in atmospheric mood and more realistic skin imperfections for a battle-hardened look. FLUX.2 [dev] is the winner for its cleaner overall composition and better clarity on the detailed textures like the leather and metalwork.

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.2 [dev]
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent layout that fully replicates a professional two-page menu structure.
  • + High-quality, distinct food photos that fit the categories.
  • + Clean and modern use of color accents and typographic hierarchy.
  • Text consists of gibberish characters rather than legible English words.
  • The price formatting is inconsistent with some very high numbers.

Seedream 4.0

  • + Perfectly legible English text for headers.
  • + Vibrant food photography with good clarity.
  • + Minimalist aesthetic using a clean white background.
  • Failed to create an actual menu layout, providing only a collage of headers and photos.
  • Lacks prices, descriptions, or any detailed menu content.
  • Irregular grid layout with awkward white space.

Verdict: FLUX.2 [dev] successfully captured the complex layout and professional design of a restaurant menu, including structured sections and stylized accents, although the text is illegible. Seedream 4.0 produced perfectly legible headers but failed completely to create a functional menu structure, resulting in a disorganized photo collage rather than a professional 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.2 [dev]
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent typography with clean, legible text and high-quality fiery effects.
  • + Highly photorealistic ingredient textures, particularly the patty and lettuce.
  • + Followed the price prompt accurately with €6.99.
  • The 'exploded' effect is a bit static and vertical compared to the dynamic request.

Seedream 4.0

  • + Stronger sense of motion with circular blurring and a more aggressive exploded layout.
  • + Great integration of the 'MAGIC BURGER' title into the flame effect.
  • Failed the pricing prompt by displaying €5.99 instead of €6.99.
  • The composition is a bit cluttered with two burgers/half-burgers, which confuses the 'exploded components' instruction.
  • The starburst graphic for the price feels less cohesive with the overall artistic style.

Verdict: FLUX.2 [dev] followed every specific prompt instruction, including the exact price and a clean exploded burger layout, while maintaining superior photorealism. Seedream 4.0 offered a more dynamic sense of motion, but failed on the price detail and created an anatomically confusing burger structure.

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.2 [dev]
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent chalk texture and smearing effects on the board.
  • + Flawless spelling and adherence to the specific pricing requested.
  • + Strong cursive title that matches the prompt's elegance.
  • One minor word repetition/overlay in the second item ('with h').

Seedream 4.0

  • + Natural, realistic chalk handwriting style.
  • + Good composition with a more dynamic angle of the chalkboard in a café setting.
  • + Consistent handwriting style across all menu items.
  • The date rendering is slightly less defined compared to Model A.
  • Slightly less chalk 'grain' texture in the letters themselves compared to Model A.

Verdict: Both models performed exceptionally well on this complex text rendering task. FLUX.2 [dev] stands out for its superior chalk texture and precise layout, while Seedream 4.0 offers a slightly more natural photographic perspective within the café. FLUX.2 [dev] is the winner for perfectly executing the exact pricing and date with high-fidelity chalk smearing details.

Pose & Character Mashup

Editing
Edit instruction

“Use Image 1 as the exact pose reference and Image 2 as the character reference. Recreate the person/character from Image 2 in the exact dynamic pose and body position from Image 1. Keep the exact face, hair, clothing style/details, and expression from Image 2. Match the lighting and environment of Image 1. The final image must show the character from Image 2 performing the precise action/pose from Image 1 with perfect anatomy and natural integration.”

Source
FLUX.2 [dev]
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Accurately captures the character's face, sunglasses, and scarf from Image 2.
  • + Matches the lighting and yellow background of Image 1 effectively.
  • Horrific anatomical failure where the original person's head from Image 1 remains attached to the side of the character's body.
  • The character is floating in a distorted way rather than standing on the ottoman.

Seedream 4.0

  • + Attempts to blend the hair and scarf into a singular figure.
  • + Maintains a cleaner background composition.
  • Fails to use the character's face, instead creating a bizarre hybrid of both people.
  • Poor anatomical rendering of the hands and feet.
  • Does not follow the vertical pose of Image 1, instead opting for a crouched position.

Verdict: Both models failed significantly on this complex editing task, but for different reasons. FLUX.2 [dev] successfully captured the character's identity from Image 2 but produced a nightmarish anatomical result by leaving the original person's head partially attached. Seedream 4.0 failed to use the correct face and ignored the specific pose instructions while also producing severe glitches in the hands and feet.

The Reversed Rodeo

Text-to-Image

“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”

FLUX.2 [dev]
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent full-body composition and lighting
  • + Realistic textures on both the horse and the spacesuit
  • + High level of cinematic detail including background planets and nebula
  • Failed the negative constraint; the astronaut is riding the horse, not the horse riding the astronaut

Seedream 4.0

  • + Strong visual clarity and dynamic posing
  • + Beautiful reflection detail in the astronaut's visor
  • + Good color balance between the white subjects and the dark space background
  • Failed the negative constraint; the astronaut is riding the horse
  • Anatomical issues where the horse's front legs merge awkwardly with the body

Verdict: Both models failed the specific spatial logic of the wordplay prompt ('horse on top, not vice versa'), instead delivering the standard 'astronaut riding a horse' trope. FLUX.1 [dev] is the superior image due to better anatomical accuracy and a more expansive, cinematic composition compared to Seedream 4.0, which has structural artifacts on the horse's chest.

Outfit Transfer Challenge

Editing
Edit instruction

“Use Image 1 as the base person. Dress them in the exact elaborate outfit from Image 2 (including all layers, accessories, jewelry, and shoes). Carefully adapt the clothing to the body shape and pose in Image 1 while maintaining realistic fabric behavior, correct proportions, and perfect lighting/shadow matching. Keep the person’s exact face, hair, and background completely unchanged.”

Source
FLUX.2 [dev]
Seedream 4.0
100% wins 0% ties 0% wins

AI Judge Analysis

FLUX.2 [dev]

  • + Successfully replicates the specific plaid scarf and navy peacoat from Image 2.
  • + High-quality rendering of fabric textures and jewelry details.
  • + Correctly maintains the background and the wooden pillar from Image 1.
  • Fails the most critical instruction by merging the faces of both people instead of keeping the person from Image 1 unchanged.
  • The vitiligo pattern and facial structure are completely altered to look like the man in Image 2.

Seedream 4.0

  • + Successfully preserves the facial identity and specific vitiligo patterns of the person in Image 1.
  • + Accurately replicates the entire outfit, including the shoes and accessories.
  • + Features better body-shape adaptation for the new clothing.
  • The scarf pattern is slightly less precise compared to the source image than Model A.
  • Minor artifacting around the sunglasses and the left hand.

Verdict: This was a complex image editing task requiring identity preservation and clothing transfer. FLUX.2 [dev] failed the primary constraint by blending the faces of both individuals, resulting in a person that no longer looks like the subject of Image 1. Seedream 4.0 successfully kept the original person's face and unique vitiligo features while accurately transferring the clothing layers, making it the clear winner for edit accuracy.

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.2 [dev]
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent photorealism with sharp textures on the capybara's fur and the businesswoman's coat.
  • + Perfect adherence to the lighting and facial expression requirements for both characters.
  • + Stronger composition that focuses on the interactions and details inside the cabin.
  • The capybara's paws look slightly more like human hands than natural capybara feet.

Seedream 4.0

  • + Natural composition showing more of the car's exterior and side profile.
  • + Good rendering of the bored expression on the passenger.
  • The capybara's paws are positioned awkwardly on the steering wheel.
  • The passenger is placed in the far back corner, making her slightly less central to the 'normalcy' of the scene than requested.
  • Slightly less clarity in the textures of the foreground elements compared to FLUX.2.

Verdict: FLUX.2 [dev] is the winner due to its superior texture rendering and more effective adherence to the 'bored' professional atmosphere requested. While Seedream 4.0 provides a good wider angle, FLUX.2 [dev] captures the humor and detail of the prompt with higher fidelity and better lighting balance.

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.2 [dev]
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent text legibility and accuracy across all fields.
  • + Clean and symmetrical composition that works well as a formal invitation.
  • + High visual clarity in the border details of thorns and webs.
  • The parchment texture is less pronounced than requested, feeling more like a digital border.
  • The jack-o'-lantern color is somewhat muted/dark compared to traditional vibrant carvings.

Seedream 4.0

  • + Atmospheric cinematic lighting with a strong glow from the jack-o'-lantern.
  • + Great vintage parchment aesthetic with torn edges and realistic textures.
  • + Dynamic composition with more depth in the background trees and bats.
  • Text on the scroll is distorted and difficult to read.
  • Text alignment at the bottom is slightly staggered and less polished.
  • Heavy shadows make some of the 'Arches' text hard to see against the background.

Verdict: FLUX.2 [dev] is the superior choice for a functional invitation because all requested text is perfectly rendered and legible, even if the atmosphere is slightly flatter. Seedream 4.0 captures the requested 'cinematic' and 'vintage parchment' vibe more effectively, but fails on the smaller text details and scroll legibility.

Bald man challenge

Image Editing
Edit instruction

“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”

Before After
FLUX.2 [dev]
Before After
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent preservation of the original face, lighting, and background.
  • + High-resolution texture with realistic individual hair strands.
  • The hairstyle choice (large afro/highly textured) may feel slightly incongruous with the existing beard texture.

Seedream 4.0

  • + Natural integration of the hairline with the forehead.
  • + The hair texture and color perfectly match the existing beard.
  • The top of the hair looks slightly flat and unnatural in its volume.
  • The lighting on the top of the hair is a bit dull compared to the strong light source in the original.

Verdict: Both models performed excellently in terms of source preservation, keeping the face, clothing, and background nearly identical to the original image. FLUX.2 [dev] provides more detailed hair texture, but Seedream 4.0 chooses a style that arguably blends more naturally with the man's existing facial hair.

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.2 [dev]
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent typography rendering and placement
  • + Realistic PBR material textures on the fish and rice
  • + Superior composition with a clean, tiered diorama base
  • The lighting is a bit soft, losing some of the 'pop' seen in the other model

Seedream 4.0

  • + Highly vibrant colors and appealing miniature aesthetic
  • + Strong adherence to the 'cartoon' aspect of the prompt
  • + Good variety of sushi types shown
  • The flag icon is placed awkwardly to the side of the text
  • The text 'SUSHI' is slightly less clean than Model A
  • The shadows are slightly harsher, losing some of the 'gentle lighting' feel

Verdict: FLUX.2 [dev] followed the prompt more precisely, particularly with the layout of the text and flag, and the implementation of a professional-looking 'raised diorama base'. While Seedream 4.0 produced a very vibrant and appealing image, FLUX.2 [dev] captured the clean, high-clarity isometric aesthetic more effectively.

Over-the-top cartoon caricature

Editing
Edit 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.2 [dev]
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent adherence to the 'caricature' style with hand-drawn line art and exaggerated head proportions.
  • + Successfully incorporates all elements: TV news desk, studio camera, multiple dogs, and a hockey rink setting.
  • + Maintains clear facial resemblance to the source image while adapting it to the art style.

Seedream 4.0

  • + Preserves the original background and clothing (denim shirt) from the source image more accurately.
  • + Successfully creates an exaggerated bobble-head style caricature of the person.
  • + Includes the hockey stick and television news desk requested.
  • Misses the 'dogs' plural requirement, featuring only one dog that looks like a standard photo insert.
  • The hand holding the phone and the hand holding the microphone have significant anatomical issues.
  • The blend between the photographic background and the caricature subject is jarring.

Verdict: FLUX.2 [dev] followed the prompt more creatively by transforming the entire scene into a cohesive caricature illustration, successfully integrating the hockey, dog, and news anchor themes into a single environment. Seedream 4.0 attempted to preserve more of the original photo's elements, but the result is a messy composite with poor hand rendering and a less imaginative interpretation of the requested career and hobbies.

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.2 [dev]
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent anatomical realism for all four animal types.
  • + Subtle, natural lighting with very clean fur textures.
  • + High level of detail in both the foreground subjects and the background meadow.
  • The scene is static and lacks the 'playfully chasing' and 'tumbling' action requested.
  • Includes an extra bunny not specified in the prompt.

Seedream 4.0

  • + Captures the 'playfully chasing' and 'tumbling' action much more effectively.
  • + Dynamic composition with animals in active poses.
  • + Beautiful use of dew sparkles and directional god rays.
  • The kitten has a slightly distorted, less realistic face compared to the others.
  • The fox's red fur is a bit oversaturated, looking less photorealistic than Image A.

Verdict: While FLUX.2 [dev] produces much higher anatomical realism and cleaner textures, it fails to capture the dynamic action of the prompt, showing the animals sitting still. Seedream 4.0 much better reflects the requested energy of animals 'tumbling' and 'chasing,' despite having slightly more 'AI-like' facial features and saturated colors.

Studio Ghibli Anime Style

Editing
Edit 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.2 [dev]
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent character likeness preservation to the source image
  • + High-quality detail in the textures and linework
  • + Perfectly captures the 'dreamy background' request with a floral field
  • The art style leans more towards modern digital anime than classic hand-painted Ghibli
  • The woman in the foreground is significantly sharper than the background elements

Seedream 4.0

  • + Perfectly captures the authentic hand-painted watercolor Ghibli aesthetic
  • + Excellent use of soft pastel colors and paper texture
  • + Faces are redesigned in a style more characteristic of Studio Ghibli
  • Lost some of the specific facial expressions/features from the original people
  • The background is very abstract compared to the structured environment of the source

Verdict: While FLUX.2 [dev] does an incredible job of maintaining the exact likeness and composition of the original meme, Seedream 4.0 captures the requested Studio Ghibli art style with much more authenticity. Seedream 4.0's use of watercolor textures and simplified character designs feels like a genuine frame from a film, whereas FLUX.2 [dev] feels like a high-detail filtered version of the original photo.

Golden Hour Stroll

Image Editing
Edit instruction

“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”

Before After
FLUX.2 [dev]
Before After
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent preservation of the source image's identity and facial details.
  • + The blowing hair is rendered with high detail and realistic movement.
  • + The flying leaves are sharp and well-integrated into the scene without obscuring the subject.
  • The wind direction appears slightly inconsistent, with hair blowing outward in both directions.

Seedream 4.0

  • + Successfully captures a sense of speed and motion through motion blur on the leaves.
  • + Introduces a warm, energetic lighting effect to enhance the 'lively' mood requested.
  • Significant loss of facial sharpness and detail compared to the source image.
  • The overall image quality appears softer and less defined than the original.

Verdict: FLUX.2 [dev] performed significantly better by perfectly preserving the subject's appearance and the background details while flawlessly integrating the requested 'wind' elements. Seedream 4.0 added a nice motion blur effect to the leaves, but at the cost of noticeable degradation in facial detail and overall image clarity.

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.2 [dev]
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent vector emblem style with clean lines
  • + Accurate spelling and inclusion of all text within a cohesive banner design
  • + Balanced minimalist composition that feels professional
  • The 'E' in 'Caffè' uses a generic grave accent rather than a stylized one

Seedream 4.0

  • + Higher quality texture on the background and logo elements
  • + More dynamic use of light and shadow on the cloche dome
  • + Accurate spelling of the name with a clear font
  • The logo is fragmented with the name floating separately from the emblem
  • Lines are slightly less crisp, looking more like an illustration than a vector logo

Verdict: FLUX.2 [dev] followed the vector emblem style more closely, creating a unified logo where the typography and icon are integrated into a single balanced mark. Seedream 4.0 has better texture and shading, but the layout is less effective as a professional logo design. FLUX.2 [dev] is the winner for its superior composition and adherence to the 'emblem style' requested.

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.2 [dev]
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev]

  • + Excellent adherence to the color palette including the specific navy and muted red.
  • + Clean, sophisticated vector icons that feel professional.
  • + Includes secondary details like the astronaut names with high legibility.
  • Layout is a grid rather than a chronological flow.
  • Text rendering becomes garbled on some labels like 'LTUREST LIAMDIIES'.

Seedream 4.0

  • + Follows the logical flow of a mission trajectory from left to right.
  • + Accurately represents icons for Earth and Moon orbits as requested.
  • + Good use of numbering to guide the eye through the steps.
  • Typography is inconsistent and includes spelling errors like 'Surfcce'.
  • Graphic design style is slightly cluttered with many competing elements.
  • Failed to properly separate steps 5 and 6 into distinct icons as requested.

Verdict: FLUX.2 [dev] produced a more aesthetically pleasing and professional-looking poster with superior vector art, though it struggled with some text labels. Seedream 4.0 better captured the chronological 'flow' of a mission infographic, but it suffered from typos and less polished graphic elements. FLUX.2 [dev] is the likely winner for its overall visual quality and adherence to the requested NASA-inspired style.

Next steps

Explore each model