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 13 shared challenges, with an AI judge weighing in on each.
Bria FIBO
#56 of 62 in Text-to-Image
GPT Image 2
#3 of 62 in Text-to-Image
Where the votes landed
Bria FIBO
0%
win rate
Ties
0%
GPT Image 2
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 refraction and light play through the glass material.
- + The plant is clearly visible through the glass as requested.
- + Clean, modern aesthetic with sharp details.
- − The blue sphere appears to be floating unnaturally rather than resting on the surface.
- − A leaf from the plant stems directly out of the top of the glass cube in a physically impossible way.
GPT Image 2
- + Natural physics, with the blue sphere resting on the bottom of the cube.
- + High-quality texture on the red book cover.
- + Accurate interpretation of 'soft window light' from the left.
- − The plant is mostly blocked by the cube and book rather than being clearly 'visible through the glass'.
- − The cube edges have a nested 'glow' effect that looks slightly digital.
Verdict: Bria FIBO captures the transparency and refraction of the glass cube more effectively, making the plant's visibility through the cube a central feature, though it suffers from some clipping issues with the plant leaves. GPT Image 2 has better physical grounding for the objects and superior texture on the book, but fails to show the plant through the glass as clearly. Overall, GPT Image 2 is the better image due to its realistic lighting and coherent composition.
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 use of reflections and wet pavement lighting
- + Strong adherence to cinematic color grading and lighting
- + Captures the mood of light rain effectively
- − The bicycle anatomy is physically impossible and nonsensical
- − Hands and fingers display Significant AI artifacts and merging
- − The subject is centered, missing the 'imperfect framing' request
GPT Image 2
- + Successfully captures 'imperfect framing' with a candid, off-center composition
- + The bicycle and tools are much more realistic and anatomically correct
- + Superior skin texture and naturalistic details on the man's face and hands
- − The lighting is a bit flat compared to the 'cinematic' request
- − Less emphasis on the wet pavement reflections than Image A
Verdict: While Bria FIBO captures a more striking cinematic atmosphere, it fails significantly on technical details, producing a bicycle and hands that are physically incoherent. GPT Image 2 follows the 'imperfect framing' and 'candid' instructions much better, providing a believable, realistic scene with far fewer structural artifacts.
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 depiction of intricate plate armor engraving
- + Strong adherence to the 'beads in hair' prompt element
- + High-contrast lighting creates a dramatic atmosphere
- − The scars on the face look somewhat artificial and flat
- − The leather and cloth textures are slightly less defined compared to the metal
GPT Image 2
- + Remarkable lifelike skin texture with natural-looking dirt and pores
- + Superior rendering of battle-worn, weathered armor with rust and scratches
- + Exquisite detail on the hair braiding and stray strands
- − Plate armor is less 'ornate' than requested in the prompt
- − The beads in the hair are less prominent than in the other model
Verdict: While Bria FIBO captures the 'ornate' aspect of the armor and the specific detail of beads more clearly, GPT Image 2 is the superior image due to its incredible photorealistic textures and lifelike eyes. GPT Image 2 much more effectively conveys the 'battle-worn' aesthetic through realistic grime and weathered equipment.
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
- + Features a bold, experimental layout with bright teal and orange accents
- + Follows the request for sections like 'Appetizers', 'Pizza', and 'Mains'
- − Text is largely illegible gibberish with inconsistent font rendering
- − Composition is cluttered and chaotic rather than clean and professional
GPT Image 2
- + Excellent text rendering with legible, realistic menu items and descriptions
- + Professional, high-quality grid layout with consistent lighting in food photography
- + Perfect adherence to the minimalist design and specified food categories
- − Layout is very traditional and safe, showing less 'artistic' risk-taking
Verdict: GPT Image 2 is significantly superior, producing a fully functional, professional-grade menu design with legible text and high-quality food photography. Bria FIBO fails on the basic tenets of graphic design for a menu, with nonsensical text and a messy, uninviting layout.
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
- + Natural and clean photorealistic rendering of the burger ingredients.
- + Successful exploded view with clear spacing between layers.
- + Accurate and legible text rendering for the main title.
- − Failed to include the Euro symbol (€) in the price.
- − The secondary text lacks the requested fiery, glowing effect.
- − Price starburst is missing the distinct geometric shape requested.
GPT Image 2
- + Perfect adherence to all text requirements, including the Euro symbol and starburst shape.
- + All text elements consistently feature the requested fiery, glowing aesthetic.
- + Superior sense of motion with sauce splashes and flying embers.
- − The composition is a bit crowded with text overlapping background elements.
- − The burger stack has slight clipping on the lettuce edges.
Verdict: While Bria FIBO produces a very clean and realistic food image, GPT Image 2 is the superior advertisement due to its perfect adherence to the complex text requirements. GPT Image 2 successfully rendered the fiery font style on all elements, included the Euro symbol, and utilized a starburst shape, while also providing more dynamic motion through sauce splashes.
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 'TODAY'S SPECIALS' text
- + Good background blur that conveys a cafe environment
- − Significant spelling errors and illegible gibberish in the menu items
- − Incorrect date (April 3 2026 instead of April 30)
- − Font style becomes inconsistent and chaotic at the bottom
GPT Image 2
- + Flawless text rendering with 100% accuracy to the prompt
- + Captures the 'handwritten-style' perfectly with subtle variations in letters
- + Excellent lighting and realistic cafe composition with a wooden frame
- − The 'gluten-free' text at the bottom is slightly small, though still legible
Verdict: GPT Image 2 is the clear winner as it followed every instruction, including specific date, prices, and complex menu item names with perfect spelling. Bria FIBO failed significantly on text rendering, producing mostly illegible text and getting the date wrong.
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 visual quality and cinematic lighting
- + Vibrant colors and a sense of movement in the horse's mane
- + Creative use of glowing hooves and reins for a sci-fi aesthetic
- − Failed the primary prompt instruction to have the horse on top
- − Interpretive only in style, not in the requested structural role reversal
GPT Image 2
- + Perfectly followed the difficult 'horse on top' instruction
- + High level of surrealism as requested
- + Detailed textures on the spacesuit and lunar surface
- − Anatomical issues where the horse's legs meet the reins
- − The horse's body feels slightly disjointed where it sits on the suit
Verdict: While Bria FIBO produced a much more aesthetically pleasing and high-quality image, it completely failed to follow the unique 'horse on top' instruction. GPT Image 2 successfully captured the surreal prompt requirements, depicting a horse literally riding an astronaut, making it the superior choice for prompt adherence despite minor anatomical artifacts.
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 depiction of the Manhattan skyline through the window.
- + Clear, high-contrast lighting that feels cinematic.
- + Captures the woman's bored expression perfectly as requested.
- − The capybara's paws look more like human hands with fur rather than actual paws.
- − The vehicle interior architecture is slightly distorted and inconsistent.
GPT Image 2
- + The capybara's paws and fur texture are much more realistic.
- + Great attention to detail on the taxi driver's cap with an 'NYC' logo.
- + The overall composition feels more like a candid, realistic photograph.
- − The passenger in the back is significantly out of focus.
- − The background lighting is a bit more generic compared to Model A.
Verdict: GPT Image 2 is the preferred choice because the capybara is anatomically more convincing, and the framing feels more like a real photograph taken from a passenger's perspective or an open door. While Bria FIBO has a more vibrant cityscape, it fails on the specific anatomy of the capybara's hands, which look somewhat unsettling.
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
- + Strong cinematic lighting on the central pumpkin
- + Follows the square format and gothic aesthetic
- − Major text hallucinations and typos in the secondary text
- − The 'frights' text is incorrectly rendered as 'nights'
- − The layout feels slightly sparse compared to the border detail
GPT Image 2
- + Perfect text rendering for all requested details including date and location
- + Highly detailed and cohesive vintage gothic composition
- + Excellent use of layers with silhouettes of 'The Arches' and a cathedral in the background
- − The parchment texture is very dark, which could make reading the small bottom text slightly more difficult
Verdict: GPT Image 2 is the superior output because it successfully renders all requested text accurately, whereas Bria FIBO suffers from significant typos and gibberish in the event details. GPT Image 2 also provides a much richer composition with detailed background elements that perfectly fit the 'The Arches, NYC' location theme.
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 soft lighting and refined PBR material textures
- + Very clean and minimalist composition
- + Perfectly followed the diorama base and solid background instructions
- − Includes some distracting nonsensical UI/watermark text in the corners
- − Sushi variety is a bit limited and basic
GPT Image 2
- + Strong text rendering with a stylized 3D effect
- + Higher complexity and detail in the miniature scene
- + Correct positioning of the flag icon between text and subject
- − Scene is very busy and slightly exceeds the 'minimal garnish' request
- − The isometric perspective is slightly inconsistent on certain objects like the bowl and lantern
Verdict: Bria FIBO captures the requested 'soft refined textures' and clean isometric aesthetic better than the competitor, resulting in a more professional miniature look. However, GPT Image 2 follows the text placement instructions more closely and avoids the messy artifact UI elements found in Bria FIBO's corners.
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
- + Excellent adherence to the list of animals requested
- + Vibrant and saturated color palette suitable for a joyful vibe
- + Clear visibility of all subjects in a balanced layout
- − The lighting and shadows feel slightly artificial and flat
- − The fox kit has a slightly cartoonish anatomy compared to the others
GPT Image 2
- + Superb photorealistic lighting with very effective god rays and backlighting
- + Dynamic and natural poses that convey the 'tumbling' and 'chasing' aspect of the prompt
- + Superior texture rendering on the fur and grass
- − The kitten's tail looks somewhat unnaturally attached or shaped
- − The kitten's paw has some minor AI artifacts
Verdict: Both models captured all four requested animals accurately, but GPT Image 2 is the superior image due to its sophisticated lighting and more natural-looking textures. While Bria FIBO has a charming, vibrant look, it leans closer to digital art than hyper-photorealism, whereas GPT Image 2 successfully captures the requested cinematic 8K masterpiece feel.
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 illustration of the cloche.
- + Uses the requested colors effectively.
- − Significant spelling errors in both the restaurant name and the date.
- − The steam effects look blurry and digital compared to the sharp lines of the logo.
GPT Image 2
- + Perfect text rendering for both the name and the requested date.
- + Excellent vintage texture and sophisticated woodcut-style illustration.
- + Stronger composition with a balanced frame and elegant typography.
- − Slightly less 'minimalist' than the other version, leaning more into ornate vintage.
Verdict: GPT Image 2 is the superior choice as it adhered perfectly to the text requirements, correctly spelling 'Caffè Florian' and '1720', whereas Bria FIBO failed both. Additionally, GPT Image 2 provided a much higher quality vintage aesthetic with better detail in the steam and texture.
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
- + Clean vector aesthetic that matches the requested flat-vector style.
- + Good use of space and navy/white muted color palette.
- − Failed to provide the numbered steps or follow the logical trajectory.
- − Text and icons are nonsensical and cluttered.
- − The Rocket (Saturn V) is inaccurate and looks more like a shuttle.
GPT Image 2
- + Excellent adherence to the specific 6-step instructions and iconography.
- + High-quality text rendering and accurate NASA branding.
- + Logical and educational layout that clearly explains the mission phases.
- − The style is a bit more detailed/illustrative than 'flat-vector' or 'simple icons'.
- − Minor detail issues like the flag on the rocket appearing slightly distorted.
Verdict: GPT Image 2 is the clear winner as it followed every specific instruction in the prompt, including the six distinct mission steps and their corresponding icons. Bria FIBO produced a visually pleasing but logically incoherent poster with garbled text and a disorganized layout that failed to convey the requested information.
Explore each model
OpenAI's state-of-the-art image generation model with arbitrary resolution up to 4K and strong instruction following