Head to head
Esc

Models · slot A

to navigate to pick

FLUX.2 [dev] Turbo fal Seedream 4.0 ByteDance

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

FLUX.2 [dev] Turbo

27.9 arena score

#3 of 62 in Text-to-Image

Top 3 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] Turbo

100.0%

win rate

Ties

0.0%

Seedream 4.0

0.0%

win rate

100.0% 0.0% ties 0.0%
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] Turbo
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent photographic realism with realistic glass textures and dust particles.
  • + Accurate physical rendering of reflections and shadows.
  • + Superior book detail, including visible spine text and realistic paper texture.
  • The glass box has an open front/side which contradicts the concept of a closed 'cube' containing an object.

Seedream 4.0

  • + Perfect adherence to the spatial 'cube' geometry with all sides closed.
  • + Dynamic lighting and caustic effects on the table surface.
  • + Clean, vibrant colors that match the prompt exactly.
  • The plant visibility through the glass is slightly muddy/blurry compared to the rest of the scene.
  • Slightly less detailed book texture than the competitor.

Verdict: Both models followed the complex spatial instructions perfectly. FLUX.2 [dev] Turbo offers a much more realistic, high-fidelity photographic look with intricate details on the book and table. However, Seedream 4.0 is the winner as it successfully rendered a closed glass cube, whereas FLUX.2 rendered an open-ended glass container, and Seedream's lighting effects on the wooden table are more visually striking.

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] Turbo
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent character preservation, maintaining the man's face, hair, and distinctive plaid coat and scarf
  • + High visual quality with realistic motion blur and lighting
  • + Maintains the specific interior details of the car, including the wood-grain dashboard
  • Changes the car color from pure white to a light silver/grey metallic

Seedream 4.0

  • + Successfully preserves the white color of the car from the source image
  • + Accurately represents the California coastline with a winding cliffside road
  • + Keeps the man's hair and scarf consistent with the source
  • The man's facial features are less accurate to the source compared to Model A
  • Perspective issues where the man appears slightly too small for the drivers seat

Verdict: FLUX.2 [dev] Turbo provides a much more convincing and high-quality edit, exceptionally preserving the man's identity and clothing. While Seedream 4.0 better preserves the original white color of the car, its character recreation is weaker and the overall composition feels more like a composite than a cohesive photo.

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] Turbo
Seedream 4.0
100% wins 0% ties 0% wins

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent adherence to all prompt elements including light rain and motion blur.
  • + Superior skin texture and high-fidelity details on the man's face and hands.
  • + Strong use of reflections on the wet pavement.
  • The bike's handlebars and front wheel geometry are slightly distorted.
  • The man's right arm appears to merge awkwardly with the bicycle frame.

Seedream 4.0

  • + Natural composition and believable candid street atmosphere.
  • + Good execution of the red bicycle color and general reflections.
  • + Motion blur on the passing vehicle is well rendered.
  • Image is noticeably softer/blurrier across the subject compared to the other model.
  • The mechanical details of the bicycle, specifically the chain and pedal area, are physically nonsensical.
  • The facial details are significantly lower quality than the competitor.

Verdict: FLUX.2 [dev] Turbo followed the prompt much more effectively, specifically capturing the skin textures and the fine details of the light rain. While Seedream 4.0 provided a nice composition, it suffered from a lack of sharpness and significant structural errors in the bicycle's anatomy. FLUX.2's output is much more cinematic and high-resolution, making it the clear winner.

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] Turbo
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Outstanding textural detail on the engraved armor and leather straps.
  • + Superior rendering of lifelike eyes and skin texture including pores and scars.
  • + Perfect adherence to the bead-braid request with many intricate small beads.
  • The lighting on the face is slightly flat compared to the dramatic backlighting.

Seedream 4.0

  • + Excellent use of warm torchlight and high-contrast dramatic lighting.
  • + Good shallow depth of field with attractive bokeh sparks.
  • + Strong facial expression that matches the 'battle-worn' theme.
  • The 'small beads' in the hair are fewer and less detailed than Image A.
  • The armor engraving and leather textures are slightly softer and less defined.
  • Minor lighting inconsistency where the torch is behind the subject but most light hits the face.

Verdict: FLUX.2 [dev] Turbo provided a significantly more detailed and technically proficient image, particularly in the rendering of the ornate armor engravings and the complex braided hair with beads. While Seedream 4.0 captured a more dramatic atmosphere with its lighting and bokeh, it lacked the intricate texture and clarity found in the leather and skin of the FLUX image.

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] Turbo
Seedream 4.0
100% wins 0% ties 0% wins

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent layout that looks like a realistic professional menu.
  • + High text legibility and mostly coherent English headings.
  • + Clearly defined sections for appetizers, pizza, and mains as requested.
  • Some placeholder text and pricing logic is slightly nonsensical ($250 for a pizza).

Seedream 4.0

  • + Features large, appetizing food photography.
  • + Follows the white background and bold sans-serif font requirements.
  • Lacks an actual menu structure; there are no item names or prices.
  • The grid layout is messy with awkward white space.
  • Missed the 'professional layout' aspect of the prompt.

Verdict: FLUX.2 [dev] Turbo successfully created a functional and aesthetically pleasing menu design with clear columns, pricing, and organized sections. Seedream 4.0 provided good quality images but failed to arrange them into an actual menu format, resulting in a disorganized collage rather than a professional document.

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] Turbo
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Perfect adherence to all text requirements including the specific €6.99 price.
  • + High photorealistic quality with clear texture on the bun and meat.
  • + Excellent use of floating sauce droplets to enhance the sense of motion.
  • The composition feels slightly static due to the central alignment.

Seedream 4.0

  • + Dynamic composition with a strong sense of rotation and movement.
  • + Vibrant fiery effects that blend well with the background.
  • Failed the negative constraint for price, showing €5.99 instead of €6.99.
  • The 'exploded' effect is less clear, as it looks like two half-burgers rather than individual components suspended.
  • Text rendering on 'LIMITED TIME ONLY' is slightly uneven.

Verdict: FLUX.2 [dev] Turbo is the superior choice because it followed every aspect of the prompt, including the specific price point and the request for individual components to be suspended. While Seedream 4.0 offered a more dynamic layout, it failed the factual text requirement and produced a less detailed breakdown of the burger ingredients.

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] Turbo
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent chalk texture with realistic smudging and dusting on the board.
  • + Flawless text rendering for every requested menu item and price.
  • + Consistent chalk stroke weight that feels truly handcrafted.
  • The layout is slightly tight at the bottom and top edges.

Seedream 4.0

  • + Very realistic composition and perspective from the side.
  • + Great natural variation in letter size and spacing as requested.
  • + Authentic-looking board with heavy chalk dust at the bottom.
  • Slightly less 'elegant cursive' style in the title compared to the other model.
  • The chalk texture on the letters is a bit thicker and less detailed than Image A.

Verdict: Both models performed exceptionally well on this complex text rendering task. FLUX.2 [dev] Turbo produced a slightly cleaner and more legible font style with highly realistic chalk grain, while Seedream 4.0 excelled in the overall cafe composition and the physical messy nature of a real chalkboard. FLUX.2 is the winner for perfectly executing every piece of specific text including the more difficult cursive elements.

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] Turbo
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Successfully integrated the man's face, sunglasses, and full outfit including the scarf.
  • + Maintained the yellow background and red ottoman from the source image.
  • Extreme anatomical failure with a second, disembodied female head attached to the shoulder.
  • Failed to replicate the specific leg crossover and foot placement of the reference pose.

Seedream 4.0

  • + Better capture of the core logic of the pose with a single character.
  • + Incorporates elements of the character's clothing like the scarf and black shirt.
  • Failed to accurately represent the face and gender of the character reference, creating a hybrid instead.
  • The arm and hand anatomy is distorted, especially the left hand and the connection to the torso.
  • Significant visual artifacts around the face and eyes.

Verdict: Both models struggled significantly with this complex pose-transfer task. FLUX.2 [dev] Turbo successfully captured the character's details but suffered a catastrophic anatomical failure by including an extra head, while Seedream 4.0 attempted a more coherent single-body integration but failed to preserve the character's identity and produced low-quality facial features. FLUX.2 is slightly preferred only because the character's appearance is correct, even though the body is grotesque.

The Reversed Rodeo

Text-to-Image

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

FLUX.2 [dev] Turbo
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + High visual clarity and cinematic space background
  • + Excellent anatomical detail on the horse and space suit
  • + Dynamic and well-balanced composition
  • Failed the specific prompt instruction to place the horse on top of the astronaut

Seedream 4.0

  • + Atmospheric lighting with nice reflections on the visor
  • + Good texture on the horse's mane
  • Failed the specific prompt instruction to place the horse on top of the astronaut
  • Noticeable anatomical artifacts with the horse's front legs and reins merging into its body

Verdict: Both models failed the negative constraint to invert the typical 'astronaut riding a horse' trope, defaulting instead to the common interpretation. FLUX.2 [dev] Turbo is the superior image due to its much cleaner anatomy, higher resolution details, and more realistic integration of the character into the cosmic environment compared to the messy leg construction in Seedream 4.0.

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] Turbo
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Successfully transferred the majority of the outfit components.
  • + Good integration of the background and lighting on the jacket.
  • Completely failed to preserve the base person from Image 1, using the face from Image 2 instead.
  • Added excessive, unrequested jewelry that was not in the source images.
  • Removed the vitiligo/unique skin patterns of the subject.

Seedream 4.0

  • + Successfully preserved the identity of the person from Image 1, including the vitiligo patterns on the forehead.
  • + Included more layers of the outfit including the shoes as requested.
  • + Maintained the correct pose and background composition.
  • The face rendering is a bit muddy and lost some clarity compared to the original.
  • The blending of the facial hair from the clothes-source onto the original person's face is slightly messy.

Verdict: The main goal was to dress the person from Image 1 in the outfit from Image 2. FLUX.2 [dev] Turbo failed significantly by replacing the person entirely with the face from the second image, whereas Seedream 4.0 successfully kept the base person's likeness and skin markings while correctly applying the clothing and shoes.

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] Turbo
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent photorealism in the capybara's fur and clothing textures.
  • + Perfect adherence to the facial expression requirements for both characters.
  • + High-quality perspective looking through the windshield from the front.
  • The passenger is sitting in the front passenger seat, failing the 'back seat' part of the prompt.
  • The 'TAX' sign on the roof is slightly mirrored or oddly rendered.

Seedream 4.0

  • + Correctly places the human passenger in the back seat as requested.
  • + Good cinematic lighting and realistic car reflections.
  • + Captures the professional posture of the capybara driver well.
  • The text on the cap is nonsensical ('TAT').
  • The passenger's face is slightly less clear and detailed than Model A.

Verdict: Both models performed exceptionally well on the surreal request, but Seedream 4.0 is the winner because it correctly placed the passenger in the back seat, whereas FLUX.2 [dev] Turbo placed her in the front next to the driver. While FLUX.2 offered slightly sharper textures on the capybara, Seedream 4.0 better captured the spatial composition of a standard taxi ride.

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] Turbo
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent typography with perfect spelling in all fields.
  • + Clean, logical composition with a well-integrated border and scroll banner.
  • + Consistent vintage illustration style throughout.
  • The parchment texture is a bit flat compared to Model B.

Seedream 4.0

  • + Dynamic cinematic lighting on the jack-o-lantern and background.
  • + High-contrast, moody atmosphere with a strong photographic feel.
  • + Good adherence to the gothic font style.
  • Text in the scroll banner is illegible and garbled.
  • Minor typography errors including a lowercase 'n' in '7pm' and messy 'Location' text.
  • The border appears a bit messy and over-processed in the corners.

Verdict: FLUX.2 [dev] Turbo produced a superior invitation because it rendered all requested text perfectly and maintained a cohesive layout. While Seedream 4.0 had more atmospheric lighting, it failed critical details like the legibility of the scroll banner and accurate text rendering for the event details.

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] Turbo
Before After
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent preservation of the original facial features and clothing
  • + Highly realistic hair texture with fine strands and natural lighting integration
  • + Natural, soft transition at the hairline
  • The requested style is a very large Afro, which might be more volume than expected for 'natural hair' in this context

Seedream 4.0

  • + Successfully adds a full head of hair that fits the head shape well
  • + Preserves the overall composition and lighting of the source
  • The hair texture appears slightly blurry or 'painted' compared to the high-frequency detail of the beard
  • The hairline is a bit too rigid and lacks the fine baby hairs seen in the other model

Verdict: FLUX.2 [dev] Turbo provided a much more sharp and detailed hair texture that perfectly matches the resolution and grain of the source image's beard. While Seedream 4.0 also followed the prompt well, its hair rendering is softer and looks more like a digital overlay than a natural part of the photograph.

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] Turbo
Seedream 4.0
100% wins 0% ties 0% wins

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent text rendering with clean, professional typography.
  • + Very high-quality PBR material rendering, especially on the salmon and rice textures.
  • + Perfect adherence to the 'diorama base' and 'isometric' perspective requirements.
  • The composition is a bit more 'realistic' than 'cartoon' compared to the other model.

Seedream 4.0

  • + Captures the '3D cartoon' style very effectively with soft, rounded shapes.
  • + Accurate text and icon placement.
  • + Good variety of sushi types within the miniature scene.
  • The text is slightly less crisp than Model A.
  • Lighting is a bit flatter, losing some of the 'refined texture' feel requested.
  • The diorama base is less distinct than the one in Model A.

Verdict: Both models followed the prompt exceptionally well, including the specific text and flag requirements. FLUX.2 [dev] Turbo produced a higher fidelity image with superior material textures and cleaner typography, while Seedream 4.0 leaned more into the 'cartoon' aesthetic with a wider variety of sushi models. FLUX.2 [dev] Turbo is the winner for its professional finish and more convincing 3D diorama presentation.

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] Turbo
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent artistic caricature style with a classic cartoon feel.
  • + Perfectly integrates all requested elements: hockey arena, anchor desk, and multiple dogs.
  • + Maintains very high facial likeness while stylizing it.
  • The text on the desk is nonsensical gibberish.
  • The composition is a bit crowded with many repeating dog elements.

Seedream 4.0

  • + Preserves the original background setting and clothing style better than the other model.
  • + Incorporates a microphone and phone to emphasize the modern media anchor role.
  • Fails to fully integrate the 'hockey' aspect beyond a small jersey logo and a stick leaning against the desk.
  • The transition between the desk and the floor is physically impossible/distorted.
  • The expression is a bit unsettling rather than humorous.

Verdict: FLUX.2 [dev] Turbo provided a much more creative and cohesive caricature by fully reimagining the scene in a professional hockey studio, whereas Seedream 4.0 simply pasted objects into the original room. FLUX.2 better captured all specific prompt requirements (hockey, dogs, anchor) in a consistent artistic style.

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] Turbo
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent prompt adherence, correctly featuring all four requested animals with distinct characteristics.
  • + Very high detail in fur texture and expressive facial features.
  • + Superior background integration with beautiful god rays and realistic dew sparkles.
  • The kitten has an anatomical oddity with a fifth leg/paw visible near the fox's head.
  • Slightly over-sharpened look on certain flowers.

Seedream 4.0

  • + Strong dynamic composition with animals appearing more 'in motion' chasing the butterflies.
  • + Beautiful lighting effects with bright, atmospheric golden hour bokeh.
  • The fox kit's anatomy is slightly distorted with a very elongated and awkwardly placed neck/chest area.
  • The dew/sparkle effects look like floating white blobs rather than realistic water droplets.
  • Finer details like the kitten's eyes are a bit less refined compared to the other model.

Verdict: FLUX.2 [dev] Turbo produced a more technically detailed image with better fur texture and environment rendering, though it suffered from a minor limb artifact. Seedream 4.0 captured the 'tumbling' motion and joyful vibe more effectively, but fell short in anatomical accuracy and realism of the dew effects. FLUX.2 [dev] Turbo is the preferred choice for its superior clarity and adherence to the 'hyper-photorealistic' part of the prompt.

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] Turbo
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent preservation of the original composition and poses.
  • + Captures the intricate illustrative style found in modern Ghibli films.
  • + Successfully incorporates dreamy background elements like cherry blossoms and soft lighting.
  • The man's facial expression is slightly more neutral compared to the original puckered lips.

Seedream 4.0

  • + Successfully captures a minimalist watercolor aesthetic reminiscent of 'The Tale of the Princess Kaguya'.
  • + Preserves the basic character silhouettes and colors.
  • Loses almost all background detail from the source image.
  • Character expressions are significantly simplified and lose the intensity of the original 'distracted boyfriend' meme dynamic.

Verdict: FLUX.2 [dev] Turbo is the clear winner as it perfectly balances the Ghibli aesthetic with the specific details of the source image, maintaining the character's facial expressions and background context. Seedream 4.0 captures a nice watercolor texture but fails to preserve the background and the nuanced emotions of the subjects.

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] Turbo
Before After
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent prompt adherence for blown hair
  • + High volume of flying leaves creates an energetic atmosphere
  • + Maintains near-perfect identity with the source image
  • The hair effect looks somewhat symmetrical and artificial
  • Dense leaves can look a bit cluttered

Seedream 4.0

  • + Natural and aesthetically pleasing hair movement
  • + Good motion blur on the flying leaves
  • + Added sunshine flare contributes to a lively feel
  • Fewer leaves than Model A makes the 'lively' effect more subtle
  • The facial features changed slightly from the original source

Verdict: FLUX.2 [dev] Turbo provided a more literal and dramatic interpretation of the 'dynamic motion' prompt, exceptionally preserving the original image's character while adding a high density of leaves. Seedream 4.0 created a more artistic and natural result with realistic motion blur, though it slightly altered the woman's face and was more conservative with the leaf count.

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] Turbo
Seedream 4.0
100% wins 0% ties 0% wins

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Perfect adherence to all text requirements including the grave accent on 'Caffè'.
  • + Excellent vector emblem style with balanced, professional composition.
  • + Subtle, realistic parchment texture that enhances the vintage feel.
  • The steam lines are slightly disconnected from the cloche.

Seedream 4.0

  • + Strong 'vintage' color palette and clean typography.
  • + Good illustration of the cloche dome with highlights and shadows.
  • + Accurate text rendering for both the name and the date.
  • Missing the grave accent on 'Caffè'.
  • The banner construction is slightly awkward with the ribbon ends disconnected from the main scroll.
  • The steam looks a bit more like generic smoke than steam.

Verdict: FLUX.2 [dev] Turbo followed the prompt more precisely, specifically including the correct accent mark on 'Caffè' and creating a more cohesive banner design. While Seedream 4.0 produced a high-quality vintage graphic, it missed the linguistic detail and had slight structural issues in the banner execution.

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] Turbo
Seedream 4.0

AI Judge Analysis

FLUX.2 [dev] Turbo

  • + Excellent text rendering with almost perfect spelling and labels.
  • + Highly detailed illustrations of the Saturn V and Lunar Module.
  • + Includes professional design touches like crew silhouettes and a NASA-inspired layout.
  • Confusing visual flow with repetitive icons for 'Translunar' and 'Earth Orbit'.
  • Literal interpretation of the prompt led to 'Saturn Vicon' being written as text instead of just showing the icon.

Seedream 4.0

  • + Very clear linear progression that follows the requested numerical steps 1-6.
  • + Clean, minimalist vector style that adheres well to 'flat-vector style' with crisp lines.
  • + Strong adherence to the requested color palette.
  • Text rendering is slightly inconsistent, with some 's' characters appearing distorted.
  • Missing icons for the final two steps (Descent and Landing), using a single image for both.

Verdict: FLUX.2 [dev] Turbo produced much higher quality individual icons and superior text rendering, though the overall layout is cluttered and logically recursive. Seedream 4.0 followed the structural instructions for a step-by-step infographic much better with a clear linear flow, but it failed to provide unique icons for the final stages and had lower-quality text. FLUX.2 is the prefered choice for its professional aesthetic and detail despite the layout messiness.

Next steps

Explore each model