Bria's 8B parameter model for controlled text-to-image generation with structured JSON prompts, trained exclusively on licensed data
Settled by community votes across 17 shared challenges, with an AI judge weighing in on each.
Bria FIBO
#56 of 62 in Text-to-Image
Seedream 4.5
#9 of 62 in Text-to-Image
Where the votes landed
Bria FIBO
0%
win rate
Ties
0%
Seedream 4.5
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
Bria FIBO
- + Strong colors and high photographic clarity
- + Excellent refraction and caustics through the glass
- + Follows all spatial instructions including lighting direction
- − The blue sphere is quite large, pushing the definition of 'small'
- − The plant appears to be growing through the cube's base rather than sitting behind it
Seedream 4.5
- + Better interpretation of 'small' regarding the blue sphere
- + Realistic glass textures and subtle reflections
- + Accurate placement of the plant behind the object
- − The glass box lacks a bottom face, appearing more like a glass cover
- − The composition is a bit tight on the top edge
Verdict: Both models followed the complex spatial instructions very well, correctly placing all elements. Bria FIBO produced a more vibrant and visually striking image with impressive light physics, while Seedream 4.5 adhered better to the relative scale of the objects (the 'small' sphere) but suffered from a slight logic error where the glass cube appears to have no bottom surface.
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
Bria FIBO
- + Strong bokeh and shallow depth of field as requested
- + Detailed skin texture on the subject's face
- + Good color saturation of the red bicycle
- − Anatomical issues with the hands, including an extra thumb and fused fingers
- − The bicycle geometry is broken, notably where the pedals and chainring connect
- − Fails to include the requested motion blur for passing cars
Seedream 4.5
- + Successfully captures motion blur of passing cars
- + More realistic 'imperfect framing' that feels like a candid street photo
- + Better integration of the rain effect on the subject's clothing and the pavement
- − The facial texture is slightly muddy and less detailed than Image A
- − The hand anatomy is still somewhat messy where the fingers touch the bicycle spokes
- − The bicycle chain and gears have some structural inconsistencies
Verdict: Seedream 4.5 more accurately followed the technical requirements of the prompt, specifically capturing the motion blur and the gritty, candid atmosphere of a rainy street. While Bria FIBO had higher facial detail, it failed on the motion blur requirement and produced significant anatomical errors in the subject's hands.
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
Bria FIBO
- + Excellent engraving and scrollwork detail on the plate armor
- + Good adherence to the bead-braided hair requirement
- + Balanced composition with clear bokeh sparks in the background
- − The scars on the face look a bit like smudged dirt rather than distinct healed tissue
- − Leather straps are less prominent in the composition
Seedream 4.5
- + Exceptional skin texture, with very realistic scars and dirt pores
- + Highly detailed rendering of the leather strap and its stitching
- + Superior lighting effects with a more natural torchlight glow on the face and metal
- − The side-profile crop makes certain armor details harder to see compared to the front view
- − Less emphasis on the ornate engraving requested in the prompt
Verdict: Both models performed exceptionally well on this complex prompt. Seedream 4.5 is the winner due to its superior lifelike skin textures and the convincing rendering of scars and grime, which better captures the 'battle-worn' aesthetic. While Bria FIBO features more intricate metallic engravings, Seedream 4.5 offers a more cinematic depth and higher fidelity in the cloth and leather materials.
Modern Clean Menu
Text-to-Image“Modern minimalist restaurant menu design, white background with colorful food photos in grid, sections for appetizers/pizza/mains, bold sans-serif fonts, vibrant accents, clean professional layout for casual dining.”
AI Judge Analysis
Bria FIBO
- + Excellent visual density with vibrant, colorful accents
- + Captures the professional complex layout of a modern menu
- + Follows the request for a grid structure and sans-serif fonts
- − Text is largely undecipherable gibberish
- − Food photography lacks distinct realism, appearing more like abstract collages
Seedream 4.5
- + High-quality, realistic food photography specifically matching 'Pizza' and 'Mains'
- + Clean, readable layout with logical price alignment
- + Accurate rendering of the requested bold sans-serif headers
- − The layout is very static and vertically stacked rather than a dynamic grid
- − Design is slightly generic and lacks professional 'branding' elements
Verdict: Seedream 4.5 is the clear winner as it provides usable, high-quality food photography that accurately matches the category headers, whereas Bria FIBO creates an abstract and cluttered design. Seedream 4.5 also maintains better text legibility and clean negative space, which is essential for a minimalist menu.
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
Bria FIBO
- + Excellent photorealistic texture on the meat patty and vegetables
- + Accurately rendered all requested text phrases
- + Highly vibrant and detailed 'exploded' burger composition
- − Missed the currency symbol (€) and starburst element for the price
- − The secondary text lacks the fiery/glowing effect of the title
Seedream 4.5
- + Perfect adherence to all text requirements including the € symbol and starburst
- + Superior sense of motion with motion-blurred ingredients flying outwards
- + Integrated fiery effect consistently across all text elements
- − The burger is less 'exploded' than Model A, appearing mostly intact
- − Texture on the lettuce and bun is slightly softer and less detailed
Verdict: Both models performed well, but Seedream 4.5 is the winner due to its superior adherence to the complex text and layout requirements, including the specific starburst price tag and currency symbol. While Bria FIBO had slightly sharper food textures, it missed several prompt elements and the burger composition felt a bit more static.
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
Bria FIBO
- + Natural-looking chalk texture on the lettering.
- + Good contrast and lighting on the chalkboard surface.
- − Failed to render the specific text requested, resulting in gibberish.
- − Incorrect date (April 3 instead of April 30).
- − Poor layout with text overlapping and trailing off at the bottom.
Seedream 4.5
- + Excellent text rendering with near-perfect spelling of long menu items.
- + Accurate adherence to the specific date and price details in the prompt.
- + Professional layout that fits all requested items clearly on the board.
- − Includes a minor repetition error on the first menu item line.
- − The handwriting looks slightly more uniform, bordering on a font-like appearance in some sections.
Verdict: Seedream 4.5 is the clear winner as it successfully rendered almost all the complex text requested in the prompt with high accuracy. Bria FIBO failed significantly on the text-to-image task, producing illegible words and incorrect information despite having a nice aesthetic chalk texture.
The Reversed Rodeo
Text-to-Image“Horse riding astronaut in space — horse on top, not vice versa. Surreal, highly detailed, cinematic.”
AI Judge Analysis
Bria FIBO
- + Crisp, clean rendering with vibrant neon accents on the hooves and reins.
- + Excellent attention to textures on the astronaut's suit and the horse's coat.
- + Strong composition with a balanced background involving a galaxy and nebulae.
- − Failed the negative constraint; the astronaut is on top of the horse, not vice versa.
- − The horse appears to be standing on an invisible ground plane rather than floating.
Seedream 4.5
- + Dynamic, cinematic pose with the horse rearing in space.
- + Naturalistic lighting that blends the subjects well into the celestial environment.
- + Detailed rendering of the astronaut's visor and equipment.
- − Failed the negative constraint; the astronaut is on top of the horse instead of the reverse.
- − Anatomical anomaly with a floating foot/stirrup visible behind the horse's belly.
Verdict: Both models failed the specific spatial logic requested in the prompt ('horse on top, not vice versa'), which was likely a trick to test adherence to unusual compositions. Between the two, Bria FIBO produces a sharper, more illustrative image with creative neon details, whereas Seedream 4.5 offers a more cinematic and painterly atmosphere, despite a glaring anatomical artifact near the center of the horse.
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
Bria FIBO
- + Excellent interior detail and lighting coherence
- + Effective use of profile and depth of field
- + Accurate depiction of the New York skyline through the window
- − The capybara's hands look more like human primate hands than capybara paws
- − The cap is more of a chauffeur style than a standard taxi cap
Seedream 4.5
- + Perfect adherence to the front-facing composition requested for the steering wheel
- + Excellent rendering of the capybara's fur and calm expression
- + Very accurate taxi driver cap with text rendering
- − The passenger's hand holding the phone is slightly distorted
- − Background bokeh is a bit generic compared to Model A
Verdict: Both models followed the prompt exceptionally well, but Seedream 4.5 is the winner for its superior rendering of the capybara's facial expression and more natural-looking paws on the steering wheel. While Bria FIBO has a more sophisticated cinematic lighting setup, Seedream 4.5 captured the specific deadpan humor and professional 'taxi driver' look more effectively.
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
Bria FIBO
- + Excellent adherence to the 'parchment' texture and thorn border requirement
- + High level of atmospheric detail with fog and fine spiderwebs
- + Accurate rendering of the main titles and banner shape
- − The event details at the bottom are garbled with significant spelling errors
- − Misquoted the text on the banner to say 'night of nights' instead of 'night of frights'
Seedream 4.5
- + Perfect text accuracy for the event details including date, time, and location
- + Clean and legible typography for the header and banner
- + Strong cinematic lighting on the central jack-o-lantern
- − Lacks the dark parchment texture requested in the prompt
- − The border is less cohesive, with thorns only appearing at the bottom and webs at the top
Verdict: While Bria FIBO captures the aesthetic of a 'vintage gothic parchment' much more effectively, it fails significantly on the text rendering for the event details. Seedream 4.5 produces perfectly legible and accurate text for all specific fields, making it the more functional invitation despite possessing a slightly more generic digital look.
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
Bria FIBO
- + Successfully added a thick, full head of hair as requested.
- + Maintains high resolution and realistic hair texture.
- − Fails to preserve source features, completely changing the man's nose, eyes, and facial structure.
- − Changes the background and clothing details significantly from the source image.
Seedream 4.5
- + Excellent source preservation, keeping the man's face, glasses, and background identical to the original.
- + Successfully adds hair that matches the existing beard color and lighting.
- + Naturally integrates the new hairline with the existing forehead features.
- − The hair density at the top is a bit thin for a 'full, thick head' of hair.
- − The transition between the top hair and the original side hair could be smoother.
Verdict: Seedream 4.5 is the clear winner because it correctly performs an image edit by preserving the original subject's identity and environment while adding the requested hair. Bria FIBO fails as an editing model, instead generating a completely new person who only vaguely resembles the original, violating the 'preserve facial features' instruction.
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
Bria FIBO
- + Excellent text rendering and placement according to prompt instructions.
- + Highly detailed food textures with distinct surfaces for rice, fish, and wasabi.
- + Clean and professional aesthetic with a perfect solid background.
- − Includes some unnecessary UI-like artifacts and watermarks at the bottom of the frame.
- − The wasabi dollops are scattered slightly haphazardly off the plate.
Seedream 4.5
- + Strong isometric composition with a clear 3D miniature diorama feel.
- + Accurate representation of requested text and small flag icon.
- + Good use of depth of field and soft lighting.
- − Texture quality is slightly muddy, particularly on the sushi rice.
- − The diorama base surface has a grainy, sandy texture that feels less refined than Model A.
Verdict: Bria FIBO provides a much higher level of detail in the food textures and maintains a cleaner overall image, though it unfortunately includes several small artifacts in the corners. Seedream 4.5 captures the miniature diorama concept well, but the text is slightly crowded at the top and the visual clarity of the sushi itself is lower than the competitor.
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
Bria FIBO
- + Expertly captures the 'exaggerated and humorous' tone of a traditional caricature
- + Successfully incorporates the news anchor and dog elements with high energy
- − Fails to include any clear hockey-related elements
- − Almost completely loses the subject's likeness from the source image
- − Background is cluttered and chaotic with minor artifacts
Seedream 4.5
- + Maintains an excellent likeness of the original person while applying caricature proportions
- + Successfully incorporates all three elements: TV anchor desk, dog, and hockey equipment
- + Retains elements of the original setting/clothing for better source preservation
- − The caricature style is very mild and less 'humorous' than requested
- − The scale of the dog versus the hockey equipment is slightly inconsistent
Verdict: Seedream 4.5 is the winner because it successfully followed all parts of the prompt, including the specific request for hockey themes which Bria FIBO missed entirely. Seedream 4.5 also did a much better job of preserving the source subject's facial features within the caricature style, whereas Bria FIBO created a generic character that bore little resemblance to the original woman.
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
Bria FIBO
- + Stronger adherence to the 'lush wildflower' request with dense floral elements
- + Clearer visibility of the fox's anatomy and characteristic tail
- + Bright and cheerful color palette
- − The fox appears to be floating unnaturally in the air without a clear jump arc
- − Anatomy of the puppy's paws is a bit simplified and lacks texture
- − Overall look leans more towards digital illustration than photorealism
Seedream 4.5
- + Excellent depiction of motion and energy with the animals running towards the camera
- + Superior light quality with distinct god rays, dew sparkles, and a soft morning haze
- + Fur textures are significantly more detailed and realistic
- − The bunny is positioned awkwardly behind the other animals
- − One of the butterfly wings on the top right is slightly clipped
Verdict: Seedream 4.5 captures the 'hyper-photorealistic' and 'wholesome vibe' much more effectively through its superior use of lighting, atmospheric effects, and convincing sense of motion. While Bria FIBO includes all the requested elements, its composition feels static and the rendering style is more reminiscent of a cartoon or bright digital painting rather than a photorealistic masterpiece.
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
Bria FIBO
- + Successfully creates a dreamy, warm atmosphere with soft lighting.
- + Captures a general anime aesthetic that is visually appealing.
- + Shows creativity in adding floral details to the background.
- − Completely loses the original composition and character poses.
- − The art style leans more towards modern shoujo manga than the specific hand-painted Studio Ghibli look.
- − Fails to preserve the source image's layout and meme-standard narrative.
Seedream 4.5
- + Excellent preservation of the source image's composition and character poses.
- + Accurately captures the specific Studio Ghibli character design and watercolor textures.
- + Perfectly translates the iconic meme into the requested artistic style while keeping it recognizable.
- − The transition between the characters and the background is a bit sharp.
- − Slightly less 'dreamy' lighting compared to Model A.
Verdict: Seedream 4.5 is the clear winner as it successfully performs the edit while maintaining the original image's character poses and composition, which are essential for this 'Distracted Boyfriend' meme. Bria FIBO completely regenerates the scene in a generic anime style, losing the iconic interaction and spatial layout of the source photo.
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
Bria FIBO
- + Excellent depiction of motion through both flying leaves and running posture
- + Successfully added dramatic wind effects to the hair
- + Captures an energetic, lively atmosphere that feels cohesive
- − Poor source preservation, essentially generating a new image that looks different from the original
- − Noticeable anatomical artifacts in the woman's hands
Seedream 4.5
- + Strong source preservation, maintaining the original woman and dog's likeness
- + Effective hair-blowing effect that feels integrated into the scene
- + Added leaves as requested while keeping the original composition intact
- − The 'lively' feel is more subtle, as the character is still walking rather than running
- − Some of the leaves appear flat or disconnected from the lighting
Verdict: Bria FIBO created a very dynamic and exciting image but failed as an editor by completely changing the character's face, clothes, and the background elements. Seedream 4.5 is the superior editor because it successfully added the requested motion, wind, and leaves while perfectly preserving the identity of the person and dog from the source image.
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
Bria FIBO
- + Clean vector-style outlines
- + Pleasant warm color palette
- + Good texture on the background
- − Significant spelling errors in both the restaurant name and the date
- − Glass effect on cloche looks more modern than vintage
- − Banner placement feels disconnected from the base
Seedream 4.5
- + Perfect text rendering for all requested words and numbers
- + Strong vintage vector aesthetic with classic shading
- + Excellent composition with arched text and balanced banner
- − Steam is a bit simple compared to the resto of the logo detail
Verdict: Seedream 4.5 is the clear winner as it followed all textual instructions perfectly, including the specific name and founding date. Bria FIBO failed significantly on the typography, misspelling the name as 'Caffé Froaian' and the date as '1320' instead of '1720'.
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
Bria FIBO
- + Successfully uses the requested color palette across the entire canvas.
- + Includes a complex, integrated layout that feels like a professional poster design.
- + Creatively interprets the icons for Earth, Moon, and the lunar module.
- − Text is entirely gibberish, failing the legibility required for an infographic.
- − The logical flow of the steps is confusing and lacks the numbered clarity requested.
- − The Saturn V icon lacks the correct silhouette, looking more like a generic shuttle.
Seedream 4.5
- + Excellent text rendering, correctly spelling NASA, names, and all six process steps.
- + Directly follows the requested six-step sequence with clear numbering and labels.
- + Features a very clean, modern vector aesthetic that perfectly matches the 'flat' style requested.
- − Missing the descent/landing icons for steps 5 and 6, replacing them with a satellite and nothing at all on the timeline.
- − The 'translunar' arc is cut off and doesn't connect the Earth and Moon as expected.
- − Composition feels a bit empty in the middle compared to the top and bottom.
Verdict: Seedream 4.5 is the clear winner because it successfully renders legible, accurate text for all mission steps and crew names, which is essential for an infographic. While Bria FIBO has a more sophisticated artistic layout, its text is complete nonsense and the sequence of information is difficult to follow. Seedream 4.5 better adheres to the numbered list requested in the prompt despite missing specific lunar module icons in the timeline section.
Explore each model
ByteDance's latest image generation model unifying text-to-image and image editing in a single architecture, with improved text rendering and 30-40% faster generation than v4.0