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
#54 of 62 in Text-to-Image
Wan 2.7
#39 of 62 in Text-to-Image
Where the votes landed
Bria FIBO
0%
win rate
Ties
0%
Wan 2.7
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 adherence to lighting and composition instructions.
- + The blue sphere is perfectly centered and looks aesthetically pleasing.
- + High clarity and clean glass reflections.
- − The plant stem appears to be growing through the bottom of the glass cube rather than behind it.
- − The blue sphere is floating without a clear reason, which may look slightly unnatural.
Wan 2.7
- + Highly realistic textures on the wooden table and the red book.
- + Convincing glass refraction and surface reflections.
- + Successfully places the plant behind the cube as requested.
- − The glass cube has strange internal structural lines that don't match a solid cube.
- − There is a duplicate blue sphere reflection on the left that feels more like a second object than a mirror image.
Verdict: Bria FIBO followed the spatial instructions better regarding the composition of the cube and sphere, creating a very clean and attractive image. However, Wan 2.7 has significantly better texture work and realism, particularly on the book and table, and correctly placed the plant behind the cube rather than appearing to sprout from it. Wan 2.7 is preferred for its superior photorealistic quality despite minor geometry issues with 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
- + Captures excellent light reflections and a moody, cinematic atmosphere.
- + Shows visible raindrops hitting the surface and subject.
- + Successfully incorporates the motion blur from background traffic requested in the prompt.
- − Significant anatomical and mechanical errors, such as a hand merging directly into the bicycle handlebars.
- − The bicycle geometry is physically impossible with the frame and drive train elements overlapping incorrectly.
- − The skin texture on the hands looks waxy and lacks the requested natural realism.
Wan 2.7
- + Outstanding realism in skin texture and clothing materials, appearing much more like a genuine photograph.
- + The bicycle structure is far more coherent and believable than the competitor.
- + Excellent composition that feels 'candid' and 'imperfectly framed' as requested, with a natural street scene depth.
- − Lacks the motion blur from passing cars that was specified in the prompt.
- − The rain effect is very subtle, bordering on invisible compared to Model A.
- − Minor warping on some background elements like the car wheels.
Verdict: Wan 2.7 is the clear winner due to its superior realism and anatomical accuracy, successfully capturing a lifelike street photography aesthetic with impressive skin textures and believable lighting. While Bria FIBO captures the cinematic atmosphere and motion blur better, it suffers from severe AI artifacts where the man's hand blends into the bicycle and the bike's frame is structurally nonsensical.
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 and intricate engraving on the armor surface.
- + Beautiful warm lighting and bokeh spark effects.
- + Effective shallow depth of field focusing on the face.
- − The 'scars' look more like surface smudges or dirt rather than physical tissue damage.
- − The beads in the hair are very small and somewhat indistinct.
Wan 2.7
- + Excellent depiction of a 'battle-worn' character with realistic scarring and textured skin.
- + Leather straps and buckles are rendered with high fidelity and realistic material properties.
- + Hair braids and beads are detailed and match the prompt perfectly.
- − The armor engravings are less complex compared to the other model.
- − The lighting is slightly more flat, despite the presence of the torch in the background.
Verdict: Both models followed the prompt well, but Wan 2.7 feels more authentic to the 'battle-worn' description with realistic skin textures and convincing scars. Bria FIBO produced a more aesthetically 'clean' image with superior armor engraving detail, but the character looks less like a veteran and more like a model in costume. Wan 2.7 is the winner for its superior storytelling through detail and material realism.
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
- + Strong adherence to the three-column grid request
- + Bold sans-serif typography is clear and prominent
- + Vibrant, high-contrast color palette
- − Nonsense text throughout the layout
- − Visual artifacts and clipping in the top-left photo blocks
- − The layout feels more like a website mobile view than a physical menu
Wan 2.7
- + Professional and realistic graphic design quality
- + Highly legible text with actual menu descriptions and prices
- + Includes all requested sections (Appetizers, Pizza, Mains) and additional realistic details like a QR code
- − Layout is a fixed vertical list rather than a broader three-column grid
- − Some minor spelling errors in item names (e.g., 'Calanfrri')
Verdict: Wan 2.7 is the clear winner as it produces a professional, functional menu with realistic food photography and legible text that makes sense in a hospitality context. Bria FIBO adheres better to the requested grid structure but fails significantly in execution, featuring fragmented imagery and entirely unintelligible placeholder text.
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 burger patty and sesame bun.
- + Strong cinematic integration of fire and embers into the atmosphere.
- + Clear and energetic glow effect on the 'MAGIC BURGER' text.
- − Missed the Euro currency symbol (€) in the price tag.
- − Price is not in a starburst as requested.
- − The layout of ingredients feels a bit cramped compared to the 'exploded' request.
Wan 2.7
- + Perfect adherence to all text elements, including the Euro symbol and starburst.
- + Superior 'exploded' dynamic composition with splashing sauces and flying seeds.
- + Creative flaming typography that integrates perfectly with the top edge.
- − The lighting on the lettuce and tomato looks slightly more illustrative/rendered than photorealistic.
- − The bottom bun of the burger is missing from the assembly.
Verdict: Wan 2.7 is the winner due to its superior adherence to the complex layout and text requirements, successfully including the starburst, all specific text strings, and the Euro symbol. While Bria FIBO has slightly more realistic food textures, its failure to include the starburst and the correct currency symbol makes it less effective as a literal interpretation of the prompt.
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
- + Excellent chalk texture and realistic handwriting variations
- + Convincing smudges and board texture that look authentic
- − Very poor text rendering with significant misspellings
- − Failed to include the date correctly, showing April 3 instead of April 30
Wan 2.7
- + Flawless text rendering with 100% accuracy to the prompt
- + Excellent composition with consistent, legible styling
- + Correctly completed the cut-off prompt item as 'Brown Butter Chocolate Chip Cookies'
- − Text looks slightly too clean, resembling a digital font rather than natural chalk
- − Minimal variation in letter size and slant despite the prompt request
Verdict: Wan 2.7 is the clear winner due to its perfect adherence to the text requirements, including accurately predicting the completion of the cut-off menu item. While Bria FIBO has a more realistic 'handwritten chalk' texture, the text is largely illegible and fails to follow the spelling and date specified in the prompt.
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 use of lighting with glowing hoof accents and bioluminescent reins
- + Strong surreal atmosphere that aligns well with the prompt
- + Clean anatomy for both the horse and the astronaut's suit
- − The composition feels slightly more compressed than Model B
Wan 2.7
- + Impressive cinematic scale with the earth and multiple galaxies in the background
- + Very realistic horse texture and lighting
- + Clear, high-resolution details on the spacesuit and equipment
- − The horse's front left leg has a slightly awkward joint angle
- − The astronaut's hands are not grasping the reins correctly
Verdict: Both models followed the core instruction perfectly, avoiding the common reversal of the subject and rider. Bria FIBO captures the 'surreal' aspect of the prompt more effectively through its creative use of light and color, whereas Wan 2.7 provides a more traditional cinematic sci-fi aesthetic with grander scale but minor anatomical flaws.
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 perspective from the back seat, capturing both characters clearly.
- + High photographic realism in the lighting and reflections on the dashboard.
- + Accurately places the passenger in the back seat as requested.
- − The capybara's hands look more like primate hands than capybara paws.
- − The car's door frame appears slightly distorted near the top left.
Wan 2.7
- + The capybara's paws have a more realistic texture and appearance for the species.
- + Sharp details on the capybara's fur and the taxi's exterior.
- − Failed the spatial prompt by putting the businesswoman in the front passenger seat instead of the back.
- − The perspective is from outside looking in, which feels less intimate than requested.
- − The car's roof and pillar geometry is inconsistent.
Verdict: Bria FIBO followed the spatial instructions much better by correctly placing the passenger in the back seat and providing a view from within the vehicle. While Wan 2.7 has good texture detail on the capybara, it failed the core compositional requirement of the prompt by placing the passenger in the front seat.
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 with glowing pumpkin effects
- + Good gothic parchment textures on the edges
- − Text rendering is very poor with numerous typos and garbled date/location details
- − The banner text mid-image is incorrect
- − Low resolution feel in the background trees and mist
Wan 2.7
- + Perfect text rendering for all prompt-specified details
- + Excellent graphic design composition with a clean, themed border
- + Highly detailed illustration style that fits the 'polished' requirement
- − Illustration style is more 'storybook' than 'cinematic vintage gothic'
- − The moon and lighting are slightly flat compared to the atmospheric lighting of the other model
Verdict: Wan 2.7 is the clear winner as it successfully rendered every piece of custom text requested, from the banner to the specific event details. While Bria FIBO captures a moodier atmosphere, its text is completely illegible and fails the primary requirement of a functional invitation.
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 thick, voluminous hair.
- + High image resolution and clarity.
- − Completely changed the person's facial features and bone structure.
- − Modified the entire background and environmental lighting.
- − Failed to preserve the source image's identity.
Wan 2.7
- + Excellent preservation of the original person's face, glasses, and expression.
- + Perfectly maintained the background and clothing from the source image.
- + Applied the hair edit seamlessly with a natural-looking hairline.
- − The hair texture is slightly soft compared to the original beard texture.
Verdict: Wan 2.7 is the clear winner as it successfully performed the image edit while keeping the source identity and background perfectly intact. Bria FIBO failed the basic requirement of source preservation, generating an entirely new person and environment that only vaguely resembles the original composition.
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 PBR material rendering with realistic light-scattering on the fish and ginger.
- + Strong adherence to the 'miniature 3D cartoon' aesthetic with smooth, tactile surfaces.
- + Perfect interpretation of the requested isometric 45° angle and diorama base.
- − Includes undesirable watermarks/UI artifacts in the bottom corners.
- − Text layout is slightly less integrated than Model B.
Wan 2.7
- + Clean, professional text rendering and graphic design elements.
- + Very high clarity with zero artifacts or watermarks.
- + Good variety of sushi types that feel balanced within the composition.
- − The textures look slightly flatter and more plastic compared to the realistic PBR materials requested.
- − The shrimp has a slightly strange, cartoonish face that might not fit the 'refined' prompt.
Verdict: Bria FIBO captures the requested 'realistic PBR materials' and 'soft refined textures' much better, creating a highly tactile 3D miniature feel, though it is marred by visible watermarks. Wan 2.7 produces a cleaner, watermark-free graphic with superior text integration, but the materials appear more like a generic 3D model rather than a high-end render. Bria FIBO is the likely winner for its superior rendering quality and texture work, which were central to the prompt.
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
- + Very detailed studio setting with cameras and lights
- + Excellent caricature style with exaggerated facial features and expression
- + High visual complexity and vibrant colors
- − Failed to incorporate hockey elements almost entirely
- − Identity preservation is weak; she looks like a different person
- − Image is cluttered with messy, nonsensical details in the foreground
Wan 2.7
- + Successfully incorporates all elements: tv anchor, dogs, and hockey (puck, rink, helmet)
- + Better preservation of the subject's facial features and hair color from the source
- + Clear and readable text in speech bubbles
- − Simple cartoon style is less 'humorous caricature' and more generic illustration
- − Text contains a spelling error ('Rolee')
Verdict: Wan 2.7 is the clear winner because it successfully integrated every part of the complex prompt (anchor, dogs, and hockey), whereas Bria FIBO completely ignored the hockey aspect. Wan 2.7 also did a much better job of making the caricature look like the woman in the source image, while Bria FIBO changed her appearance significantly.
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
- + Successfully includes all four requested animals
- + Vibrant colors and high-contrast lighting creating a joyful mood
- + Strong character expressions with large, twinkling eyes
- − Anatomical issues with the fox's anatomy and paw placement
- − The animals appear somewhat 'cut-and-pasted' into the scene rather than integrated
- − The lighting on the animals is inconsistent with the god rays in the background
Wan 2.7
- + Excellent integration of animals into the long grass and dew
- + Superior realistic fur texture and lighting rendering
- + Natural interaction between the kitten and the bunny
- − The fox is partially cropped and lacks the playful 'tumbling' energy of the others
- − Colors are slightly more muted than the 'joyful vibe' might imply
- − The butterfly anatomy is a bit simplified
Verdict: Wan 2.7 is the winner due to its significantly higher level of photorealism and better spatial integration of the animals into the meadow. While Bria FIBO captured the requested energy well, it suffered from anatomical distortions and an illustrative quality that missed the 'hyper-photorealistic' part of the prompt.
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 captures the Ghibli aesthetic with soft lighting and floral backgrounds
- + Retains the dynamic 'distracted boyfriend' meme composition while stylizing the characters
- + Excellent use of a warm, nostalgic color palette
- − Significantly alters the characters' faces beyond simple stylization, losing the likeness of the original meme figures
- − The character on the right is walking forward instead of sideways, changing the pose's intent
Wan 2.7
- + Exceptional preservation of the original image's pose, facial expressions, and overall structure
- + High accuracy in capturing the specific identities and 'memeable' essence of the source figures
- + Beautiful watercolor hand-painted texture applied across the entire image
- − The color palette is slightly washed out compared to the vibrant 'Ghibli' request
- − The background remains very close to the original photo rather than becoming a 'dreamy' illustration
Verdict: While Bria FIBO creates a more traditionally 'pretty' anime illustration, it sacrifices the iconic nature of the original meme by genericizing the characters. Wan 2.7 is the superior edit because it perfectly translates the source image's geometry and specific facial expressions into an illustration style without losing the context of the original 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 interpretation of 'energetic and lively' by changing the character's pose to running
- + Highly dynamic hair and tail movement
- + Artistic use of falling leaves that enhances the autumn theme
- − Poor source preservation, essentially generating a new image with a similar subject
- − Face and background details differ significantly from the original
Wan 2.7
- + Perfect preservation of the source image's identity, background, and character details
- + Subtle but effective wind effect applied to the hair
- + Accurately adds falling leaves as requested without disturbing the base image
- − The motion feels less 'energetic' and more static compared to a running pose
- − Leaves are somewhat sparse and less integrated into the depth of the scene
Verdict: Bria FIBO created a more energetic and visually exciting image, but it failed the core task of 'image editing' by completely regenerating the scene and changing the woman's pose and features. Wan 2.7 successfully added the requested effects (blowing hair and leaves) while perfectly preserving the identity of the original photo. Wan 2.7 is the winner for its superior source preservation and accurate application of edits.
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 execution
- + The cloche illustration is well-rendered with subtle steam
- − Major spelling error: 'Caffé Froaian' instead of 'Caffè Florian'
- − Incorrect date: 'Est. 1320' instead of 1720
- − The banner and text alignment feels slightly disconnected
Wan 2.7
- + Successfully rendered the correct date: 'Est. 1720'
- + Rich, cohesive emblem-style composition
- + Includes the frame and extra decorative elements that enhance the vintage feel
- − Minor spelling error: 'Florion' instead of 'Florian'
- − The steam is rendered as abstract swirls rather than realistic vapor
Verdict: Wan 2.7 is the clear winner as it correctly followed the date instructions and produced a much more sophisticated emblem design. While both models struggled slightly with the typography of 'Florian', Bria FIBO failed significantly on both the brand name and the specific historical date requested.
Apollo 11: Journey to Tranquility
Text-to-Image“Create a clean, modern vector infographic poster about the Apollo 11 mission. NASA-inspired palette (navy, white, muted red, light gray). Flat-vector style, crisp lines, consistent iconography, subtle gradients only. Steps (stop at landing): 1. Launch (Saturn Vicon) 2. Earth Orbit (Earth + orbit ring icon) 3. Translunar (trajectory arc icon) 4. Lunar Orbit (Moon + orbit ring icon) 5. Descent (lunar module descending icon) 6. Landing (lunar module on the surface icon) Small supporting elements (minimal text): • Crew strip: three silhouette icons with only last names: Armstrong, Aldrin, Collins. • Landing site marker: Moon pin labeled "Tranquility" only. Layout constraints: generous margins, large readable labels, clean background with subtle stars. Vector-only, print-poster look, high resolution.”
AI Judge Analysis
Bria FIBO
- + Stronger vector aesthetic with depth through subtle drop shadows
- + More creative layout of the journey trajectory
- − Text is entirely illegible gibberish
- − The Saturn V icon looks more like a shuttle and lacks technical accuracy
- − Logic of the infographic steps is circular and confusing
Wan 2.7
- + Excellent prompt adherence following all 6 requested steps in order
- + Remarkably legible text for an AI model, with accurate names and dates
- + Authentic infographic layout that successfully conveys scale and chronological order
- − Minor spelling errors such as 'DESCRIPT' and 'Tranquiliry'
- − The star field is a bit cluttered compared to the clean vector request
Verdict: Wan 2.7 is the clear winner as it successfully follows the complex 6-step chronological instruction and produces highly legible, relevant text. While Bria FIBO has a nice aesthetic, it fails to deliver on the specific infographic content requested and contains only placeholder-style nonsensical text.
Explore each model
Alibaba's Wan 2.7 image generation and editing model for text-to-image, reference-guided generation, and instruction-based image edits