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
#55 of 62 in Text-to-Image
Z-Image Turbo
#12 of 62 in Text-to-Image
Where the votes landed
Bria FIBO
0%
win rate
Ties
0%
Z-Image Turbo
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
- + Excellent depiction of the plant visible through the glass refraction.
- + Strong visual quality with sharp details and realistic lighting on the table surface.
- + Perfect adherence to the placement of the sphere floating inside the cube.
- − The plant appears to be growing out of the cube rather than being behind it.
Z-Image Turbo
- + Accurately places the plant in the background as requested.
- + Realistic texture on the red book cover.
- − The plant is not visible through the glass cube, failing a specific part of the prompt.
- − The 'glass' cube has a solid mirror base that wasn't requested.
- − The sphere is sitting on the bottom rather than being 'inside' in a visually interesting way.
Verdict: Bria FIBO followed the spatial instructions much better, specifically the requirement for the plant to be visible through the glass. While Z-Image Turbo placed the plant in the background, it failed to render it through the cube's transparency, and added an unsolicited mirror effect to the bottom of the cube.
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
- + Excellent shallow depth of field and bokeh effects
- + Captures the 'cinematic' and 'candid' mood very effectively
- + Natural skin texture and believable elderly features
- − Significant anatomical errors with the hands (multi-fingered and merged textures)
- − The bicycle geometry is warped and physically impossible in several places
Z-Image Turbo
- + Better representation of rain streaks on the pavement and in the air
- + Realistic street scene composition including traffic
- + More accurate bicycle structure despite some minor pedal issues
- − Lacks the shallow depth of field requested, appearing very sharp throughout
- − Missing the requested 'motion blur from passing cars' as the background cars are static
- − Visual quality is slightly lower with more digital noise
Verdict: Bria FIBO captures the requested cinematic atmosphere and shallow depth of field perfectly, but fails significantly on fine details like hands and bicycle mechanics. Z-Image Turbo provides a more grounded and realistic street scene with better rain effects, though it ignores the specific request for motion blur and shallow focus. Bria FIBO is the preferred choice for its superior artistic composition despite its structural flaws.
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
- + Exquisite detail on the engraved metal scrollwork
- + Strong adherence to the 'hair braided with small beads' requirement
- + Excellent warm lighting consistency across the face and armor
- − The scars look more like surface dirt or smudges rather than healed tissue
- − The background bokeh sparks are a bit static and uniform
Z-Image Turbo
- + Realistic skin texture with more convincing depictions of scars and injury
- + Includes the physical torch as a light source, creating dynamic rim lighting
- + Excellent representation of the chainmail and cloth underlayers
- − The 'braided' hair is a bit messy and less defined than requested
- − Slightly less 'close' as a portrait compared to Model A
Verdict: Bria FIBO captures the technical details of the armor and hair accessories with incredible precision, providing a very clean and professional portrait. However, Z-Image Turbo feels more 'battle-worn' and lifelike, with superior skin textures and a more atmospheric use of lighting derived from the actual torch in the frame. Both models followed the prompt excellently, but Z-Image Turbo wins on overall realism and character depth.
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
- + Effective use of vibrant color accents in the layout corners
- + Good variety of heading sizes for hierarchical clarity
- + Clean vertical alignment for a modern brochure-style menu
- − Text is largely illegible gibberish
- − The grid layout feels slightly chaotic and cramped
Z-Image Turbo
- + Higher quality food photography that looks appetizing and professional
- + Superior font rendering with legible characters and prices
- + Very clean, balanced grid composition that adheres strictly to the prompt
- − Typographical error in 'Pizza Mans' instead of 'Pizza/Mains'
- − Slightly less creative with accent colors compared to the other model
Verdict: Z-Image Turbo is the clear winner as it produces a professional, functional menu layout with high-quality photography and legible text. Bria FIBO captures the 'vibrant accents' well, but fails to deliver realistic food imagery or readable typography.
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 adherence to the 'exploded burger' requirement with clearly separated layers.
- + Authentic fiery glow and lighting on the text and secondary elements.
- + Highly photorealistic textures on the lettuce, patty, and tomatoes.
- − The '€' symbol is missing from the price.
- − The background fire is a bit cluttered, potentially distracting from the product.
Z-Image Turbo
- + Perfect text rendering including the '€' symbol and starburst shape.
- + Very clean and professional advertising layout with centered typography.
- + Good use of depth of field with the surface and floating embers.
- − Failed the 'exploded burger' requirement, showing a mostly assembled burger instead.
- − The burger patties have some unnatural dark spots/artifacts.
Verdict: Bria FIBO followed the spatial instructions much better by creating a truly 'exploded' burger with floating layers, whereas Z-Image Turbo produced a standard stacked burger. However, Z-Image Turbo was more accurate with the text requirements, including the specific currency symbol and starburst container. Bria FIBO is the overall winner for capturing the requested dynamic motion and complex composition.
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
- + Features a more authentic chalk texture on the board surface
- + Captures an organic, uneven handwriting style
- − Failed significantly on text rendering with gibberish words
- − Did not follow the specific content or date in the prompt
- − Text is crowded and poorly laid out
Z-Image Turbo
- + Excellent adherence to the requested text and pricing
- + Very legible handwriting that maintains a realistic chalk appearance
- + Clean layout with accurate spelling despite a minor typo in 'Mushrooms'
- − Minor spelling error in 'Mustroom'
- − Slightly less realistic chalk texture compared to Model A
Verdict: Z-Image Turbo successfully rendered almost all the requested text correctly, including the specific date and menu prices, while maintaining the requested handheld chalk style. Bria FIBO failed to produce legible words for the menu items, resulting in nonsensical text throughout the image. Z-Image Turbo is the clear winner for follow-through on complex prompt instructions.
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
- + Excellent vibrant colors and cinematic lighting that fits the 'surreal' prompt.
- + Creative use of glow on the hooves and reins.
- + High-quality texture on the horse's coat and space suit.
- − The horse's front legs are poorly rendered with a strange joint structure near the chest.
Z-Image Turbo
- + Natural and realistic anatomical proportions for both the horse and the astronaut.
- + Clean, professional-looking details on the space suit and saddle.
- − Failed the specific prompt instruction 'horse on top, not vice versa' by rendering a standard rider configuration.
- − The background is very plain and lacks the 'highly detailed' space environment requested.
Verdict: Both models failed the negative constraint to have the horse on top of the astronaut, but Bria FIBO produced a much more visually compelling and 'surreal' image that captured the spirit of the prompt's aesthetic. Z-Image Turbo created a more anatomically accurate image, but it was far too conventional and lacked the cinematic detail requested.
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
- + Strong composition showing the complete interior and exterior environment
- + Excellent lighting and color contrast between the yellow taxi and the night city
- + Highly detailed capybar fur and professional human expression
- − The capybara's hands look more like human-monkey hybrids than capybara paws
- − The steering wheel placement is slightly awkward in relation to the dash
Z-Image Turbo
- + Natural look to the capybara's fur and face
- + Effective 'bored' expression on the passenger in the background
- + Good focus on the subject with nice depth of field
- − The passenger appears to be in the front passenger seat rather than the back seat
- − The lighting is less cinematic and feels somewhat flat compared to the prompt's potential
- − Missing the vibrant night city lights atmosphere through the window
Verdict: Bria FIBO captures the prompt much more effectively by placing the passenger in the back seat and creating a rich, photorealistic New York night atmosphere. While Z-Image Turbo has a nice render of the capybara, it fails the spatial requirement of having the passenger in the back and lacks the environmental detail seen in Bria FIBO.
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 handle on atmospheric lighting and fog effects.
- + Accurate spooky aesthetic with thorn and web border integration.
- + High-quality central jack-o-lantern rendering.
- − Text rendering is very poor with numerous typos and hallucinations.
- − The secondary scroll text says 'night of nights' instead of 'frights'.
- − The bottom event details are illegible and incorrect.
Z-Image Turbo
- + Exceptional text legibility and accuracy for almost all fields.
- + Distinct parchment texture that fits the vintage request.
- + Clearer scroll banner implementation for the secondary text.
- − Small typo in 'The Archves' (extra 'v').
- − The border is a bit cluttered with the white spiderwebs overlapping the thorn assets somewhat harshly.
- − The central pumpkin lighting is slightly flatter compared to Image A.
Verdict: While Bria FIBO captures a more cinematic and moody atmosphere, it fails significantly on the textual requirements, producing incoherent words. Z-Image Turbo captures nearly all the text accurately, including the specific date and time, although it has a single letter typo in 'Arches'. Z-Image Turbo is the clear winner for an invitation task where text is a primary requirement.
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 full and thick head of hair as requested.
- + Realistic hair texture and natural integration with the lighting of the scene.
- − Significantly altered the facial features, making the person appear much younger and different from the source.
- − Changed the landscape and specific details of the clothing, failing to preserve the source image context.
Z-Image Turbo
- + Near-perfect preservation of facial features, lighting, and background elements.
- + Maintains the identity of the person from the source image.
- − Failed to provide a 'full, thick head of hair', only adding a very short buzz cut/stubble.
- − Did not meet the primary instruction of the edit regarding hair density and style.
Verdict: Bria FIBO followed the creative instruction to add a full head of hair but failed the editing constraint of preserving facial features and the source environment. Conversely, Z-Image Turbo preserved the source image perfectly but failed to apply the requested edit, only adding minimal stubble instead of thick hair. Bria FIBO is the likely winner for successfully executing the core transformation, despite the loss of original identity.
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 alignment for 'JAPAN' and 'SUSHI'.
- + High-quality PBR textures with realistic subsurface scattering on the ginger and fish.
- + Includes a correct Japanese flag icon as requested.
- − Includes distracting AI watermarks and gibberish text in the corners.
- − The diorama base is a bit large, pushing the sushi assortment lower in the frame.
Z-Image Turbo
- + Perfectly centered composition and clean isometric perspective.
- + Strong 3D cartoon aesthetic with soft, appealing lighting.
- − Displays the flag of China instead of the flag of Japan.
- − Text rendering is slightly less crisp than Model A.
- − Highly simplistic scene with only one piece of sushi, lacking the 'miniature scene' feel.
Verdict: Bria FIBO followed the prompt much more accurately, providing the correct flag and a detailed 'scene' of sushi with superior material textures. Z-Image Turbo failed a key geographical instruction by placing a Chinese flag next to the word 'JAPAN' and produced a much more basic image.
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
- + Successfully applied the caricature style with exaggerated features.
- + Includes all requested elements: tv show anchor set, a large dog, and hockey sticks/equipment.
- + High creativity in the busy studio setting and colorful outfit.
- − The facial likeness to the original subject is significantly altered.
- − The hands and microphone rendering have notable structural errors.
Z-Image Turbo
- + Excellent preservation of the original subject's facial features and clothing.
- + Added a small, subtle dog to the background.
- − Completely failed to apply the caricature style or the 'exaggerated and humorous' tone.
- − Ignored the hockey and tv show anchor profession elements entirely.
- − Hardly qualifies as an edit based on the specific instructions.
Verdict: Bria FIBO followed the creative prompt instructions much more effectively by transforming the image into a stylized caricature that includes the desk, hockey equipment, and a dog, despite some anatomical artifacts. Z-Image Turbo essentially returned the source image with a tiny, blurry addition, failing to address the core requirements of profession or style. Bria FIBO is the clear winner for actually performing the requested task.
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
- + Includes all four requested animals clearly visible.
- + Vibrant and colorful composition with a high density of wildflowers.
- + Strong lighting effect with visible god rays as requested.
- − The fox looks slightly levitated rather than jumping naturally.
- − More of a digital illustration style than hyper-photorealistic.
- − Anatomical issues with the rabbit's ears and the kitten's limb placement.
Z-Image Turbo
- + Excellent soft fur textures that feel more realistic.
- + Captured the 'tumbling together' interaction much better with animals touching.
- + More natural depth of field and soft morning light.
- − The kitten and fox are partially overlapping, making their forms slightly less distinct.
- − Fewer butterflies and less variety in flowers compared to the other model.
- − The kitten's facial expression is slightly distorted.
Verdict: Z-Image Turbo is the winner for its superior texture rendering and more natural interaction between the animals, which feels much closer to a photograph. Bria FIBO followed the prompt's layout well but resulted in a more synthetic, over-saturated look that felt like a composite illustration rather than a hyper-photorealistic 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
Bria FIBO
- + Strongly captures the Studio Ghibli aesthetic with soft lighting and hand-painted textures.
- + Preserves the iconic composition and poses of the 'distracted boyfriend' meme perfectly.
- + Successfully translates the color palette into warm, nostalgic tones.
- − The characters are completely redrawn into a generic anime style, losing the likeness of the original people.
Z-Image Turbo
- + Maintains the high resolution and photographic quality of the original source image.
- − Completely fails the edit instruction by producing a photographic result instead of an illustration.
- − The facial expressions from the original source are altered, losing the specific emotion of the meme.
- − Lacks the requested pastel colors, textures, and dreamy atmosphere.
Verdict: Bria FIBO successfully executed the creative brief by transforming the meme into a beautiful Studio Ghibli-inspired world while maintaining the core composition. Z-Image Turbo failed to apply the requested artistic style, producing a near-identical photograph that ignored the 'illustration' and 'dreamy background' requirements.
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 adherence to the 'energetic and lively' request.
- + Highly dynamic hair motion and clear falling leaf additions.
- + Captures the intent of an action shot by repositioning the limbs for a running motion.
- − Low source preservation as it completely regenerates the person and dog's facial features and poses.
- − Noticeable anatomy issues with the woman's hands and the leash attachment.
Z-Image Turbo
- + Excellent source preservation, keeping the original faces and clothing nearly identical.
- + Subtle but effective hair movement that looks natural.
- + Maintains the high resolution and clarity of the original background.
- − The 'energetic and lively' feel is very understated compared to the request.
- − The leaves are small and feel like a minor overlay rather than a dynamic part of the scene.
Verdict: Bria FIBO followed the creative instructions much more aggressively, creating a genuinely dynamic scene with significant motion, but at the cost of changing the source image's identity almost entirely. Z-Image Turbo respected the source image perfectly but failed to truly deliver on the 'energetic and lively' request, providing only very subtle edits. Bria FIBO is the winner for better fulfilling the specific edit instructions regarding motion and energy.
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
- + Successfully included the banner element requested in the prompt.
- + Captures a vintage aesthetic with layered line work.
- − Significant spelling errors in the brand name ('Froaian' instead of 'Florian').
- − Incorrect year in the banner ('1320' instead of '1720').
- − Graphical issues with 'steam' appearing inside a glass dome rather than rising from it.
Z-Image Turbo
- + Perfect text rendering for both the brand name and the establishment date.
- + Clean, minimalist vector style that adheres well to modern logo standards.
- + Correct interpretation of 'steam' rising from the cloche.
- − Missed the 'banner' element for the 'Est. 1720' text.
- − The cloche icon is somewhat generic compared to the detailed line work in Model A.
Verdict: Z-Image Turbo is the clear winner because it correctly spelled the brand name and the establishment date, which are critical for a logo. While Bria FIBO attempted more of the complex prompt elements like the banner, it failed significantly on text accuracy and provided an incorrect year.
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
- + Excellent adherence to the 'NASA-inspired palette' with deep navy and muted tones.
- + Highly professional layout and composition that truly feels like an infographic poster.
- + Clean vector aesthetic with sophisticated usage of icons and supporting details.
- − The numerical 'steps' requested are not clearly delineated or sequenced.
- − The text is entirely illegible gibberish.
Z-Image Turbo
- + Text is partially legible, including 'Earth Orbit' and 'Tranquility'.
- + Follows the vector style requested with distinct icons for the moon and lunar module.
- − Lacks the requested color palette, using a bright orange instead of muted red.
- − Poor layout with icons floating disconnectedly rather than forming a cohesive infographic.
- − Contains significant spelling errors in large text ('APOLIO E 11', 'Translurian').
Verdict: Bria FIBO captures the aesthetic and professional feel of a modern infographic much better than Z-Image Turbo, utilizing a superior color palette and layout. While Z-Image Turbo attempts legible text, its spelling errors and disjointed composition make it less effective as a poster. Bria FIBO is the preferred choice for its visual coherence and adherence to the specified design style.
Explore each model
Tongyi-MAI's 6-billion parameter distilled text-to-image model optimized for speed, achieving high-quality generation in 8 steps or fewer with support for bilingual text rendering