Head to head
Esc

Models · slot A

to navigate to pick

Stable Diffusion 3.5 Large Stability AI Vidu Q2 ShengShu Technology

Settled by community votes across 13 shared challenges, with an AI judge weighing in on each.

Stable Diffusion 3.5 Large

22.9 arena score

#28 of 62 in Text-to-Image

Skill signature · Text-to-Image

Vidu Q2

19.8 arena score

#42 of 62 in Text-to-Image

Vote tally

Where the votes landed

Stable Diffusion 3.5 Large

0%

win rate

Ties

0%

Vidu Q2

0%

win rate

Shared challenges 13

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.”

Stable Diffusion 3.5 Large
Vidu Q2

AI Judge Analysis

Stable Diffusion 3.5 Large

  • + Excellent sense of realism in textures and reflections
  • + Very high resolution and photographic quality
  • Failed the spatial relational logic: the cube is over the book, not the book on top of the cube
  • The plant is in the background but not clearly 'behind the cube' in a way that shows through it effectively

Vidu Q2

  • + Perfect adherence to complex spatial instructions
  • + Realistic lighting and shadows including the refraction through the glass
  • + Correct placement of all objects (book on top, sphere inside, plant behind)
  • The glass edges are unusually thick for a standard 'glass cube'
  • The blue sphere has a slightly matte finish compared to the highlights on the glass

Verdict: Stable Diffusion 3.5 Large failed a key spatial prompt requirement by placing the glass cube on top of the red book. Vidu Q2 followed every instruction perfectly, correctly positioning the book on top of the cube and the plant behind it, while maintaining high visual fidelity and convincing shadows.

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.”

Stable Diffusion 3.5 Large
Vidu Q2

AI Judge Analysis

Stable Diffusion 3.5 Large

  • + Successfully captures the light rain and atmospheric reflections on the pavement.
  • + Includes a bus and car which provide a sense of scale and street atmosphere.
  • + Good full-body composition showing the interaction between the man and the bike.
  • The hands and the contact point between the man and the bike are blurry and poorly defined.
  • Lacks the 'motion blur' on passing cars requested in the prompt.
  • The bicycle's anatomy is slightly simplified and warped in places.

Vidu Q2

  • + Excellent close-up detail on the skin texture and used/worn metal of the bicycle.
  • + Great adherence to the 'imperfect framing' and '50mm' feel with a more intimate, candid perspective.
  • + The bicycle mechanics (chain, gears, frame) are rendered with high realism and correct positioning.
  • The passing car in the background is static and lacks the requested motion blur.
  • The rain is less visible compared to Model A, appearing more like 'post-rain' wetness than active falling rain.
  • The composition is very tight, cutting off much of the subject and the bike.

Verdict: Stable Diffusion 3.5 Large creates a more evocative atmospheric scene with better rain effects and a wider view, but the subject's anatomy is weak. Vidu Q2 provides superior detail in skin texture and mechanical accuracy, capturing the 'candid' and 'natural' part of the prompt much more effectively. Vidu Q2 is the preferred choice for its photographic realism and convincing details, despite both models failing to generate the requested motion blur.

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.”

Stable Diffusion 3.5 Large
Vidu Q2

AI Judge Analysis

Stable Diffusion 3.5 Large

  • + Excellent intricate engraving on the plate armor
  • + Realistically textured, battle-worn skin with believable dirt and faint scuffs
  • + Strong atmospheric lighting with effective shallow depth of field
  • Missed the request for beads in the hair
  • The hair texture looks slightly matted compared to the high-quality armor

Vidu Q2

  • + Successfully included beads in the braided hair
  • + Excellent lighting contrast and warm torchlight reflections
  • + Very high level of detail on leather straps and cloth underlayers
  • The face looks a bit too clean and 'pretty' for a battle-worn paladin
  • Armor engravings are less intricate than the other model

Verdict: Both models followed the prompt well, but Stable Diffusion 3.5 Large captured a more authentic 'battle-worn' aesthetic with superior skin textures and highly detailed armor engravings. While Vidu Q2 succeeded in including the beads in the hair and produced excellent leather textures, the subject's face appears too polished for the intended theme.

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.”

Stable Diffusion 3.5 Large
Vidu Q2

AI Judge Analysis

Stable Diffusion 3.5 Large

  • + Excellent photographic quality and realism in food imagery
  • + Clean and professional minimalist layout that feels like a real editorial design
  • + Clear grid structure as requested in the prompt
  • Text rendering is mostly gibberish despite high-quality fonts
  • Text alignment becomes slightly cluttered at the bottom of the central column

Vidu Q2

  • + Good use of vibrant color accents and graphical elements
  • + Logical organizational structure with distinct columns and clear prices
  • + Better grouping of specific menu sections like 'Mains'
  • Lower image quality with noticeable AI artifacts and blurry textures
  • Food photos are less appetizing and have repetitive shapes
  • The layout feels more cluttered and less minimalist

Verdict: Stable Diffusion 3.5 Large wins due to its superior visual quality and adherence to the 'modern minimalist' aesthetic. Vidu Q2 does a better job of organizing a functional menu structure with pricing, but the low resolution and artifacts make the food look unappealing compared to Stable Diffusion 3.5 Large's crisp, professional photography.

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.”

Stable Diffusion 3.5 Large
Vidu Q2

AI Judge Analysis

Stable Diffusion 3.5 Large

  • + Excellent photorealistic rendering of food textures.
  • + Dynamic lighting and atmospheric bokeh effects.
  • Failed to follow any of the text integration instructions.
  • The burger is stacked rather than exploded or suspended in mid-air.

Vidu Q2

  • + Perfect adherence to text instructions with correct spelling and glow effects.
  • + Accurately represents the 'exploded' and 'suspended' burger components.
  • + Includes the specific starburst price tag requested.
  • The price currency symbol is a stylized mashup rather than a clean euro sign.
  • Slightly less realism in the lettuce texture compared to the competitor.

Verdict: While Stable Diffusion 3.5 Large produces a more photorealistic image of a burger, it failed to follow almost all of the specific layout and text instructions. Vidu Q2 captured every element of the prompt, including the exploded composition and all three specific text strings, making it the superior choice for a creative advertisement brief.

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.”

Stable Diffusion 3.5 Large
Vidu Q2

AI Judge Analysis

Stable Diffusion 3.5 Large

  • + Provides a complete environmental context with a cozy café setting.
  • + Text layout is neat and visually organized on the board.
  • Failed the date prompt, rendering '2024' instead of '2026'.
  • Text is largely gibberish and does not follow the requested menu items.
  • Title contains a typo ('TODAAY').

Vidu Q2

  • + Excellent adherence to the 'handwritten chalk' texture and style.
  • + Followed the specific date 'APRIL 30, 2026' perfectly.
  • + Attempts all specific menu items with recognizable (though misspelled) text.
  • Numerous spelling errors throughout the menu items.
  • Lacks the wider café environment, focusing only on the board.

Verdict: Stable Diffusion 3.5 Large provides a much better overall composition of a café, but fails significantly on the specific text requirements and the requested date. Vidu Q2 captures the requested chalk texture and specific details like the date and menu items much more accurately, despite some spelling issues.

The Reversed Rodeo

Text-to-Image

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

Stable Diffusion 3.5 Large
Vidu Q2

AI Judge Analysis

Stable Diffusion 3.5 Large

  • + Excellent cinematic lighting and atmosphere
  • + High level of detail in the astronaut suit and horse texture
  • + Beautiful celestial composition with a sense of motion
  • Failed the core prompt instruction of 'horse on top' of the astronaut

Vidu Q2

  • + Vibrant colors and creative 'nebula' texture on the horse
  • + Sharp focus and clean rendering of the characters
  • + Good interpretation of the 'surreal' aesthetic
  • Failed the specific spatial instruction to place the horse on top of the astronaut
  • Prohibitive anatomical issues with the astronaut's legs passing through the horse

Verdict: Both Stable Diffusion 3.5 Large and Vidu Q2 completely ignored the specific prompt instruction to place the horse on top of the astronaut, instead providing the standard 'astronaut riding a horse' trope. Stable Diffusion 3.5 Large is visually superior due to its cinematic lighting and realistic integration, whereas Vidu Q2 has significant clipping issues where the astronaut's body intersects the horse's back.

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.”

Stable Diffusion 3.5 Large
Vidu Q2

AI Judge Analysis

Stable Diffusion 3.5 Large

  • + Excellent texture on the capybara's fur and whiskers
  • + High resolution foreground detail with realistic lighting
  • Completely failed to include the passenger in the back seat
  • One paw is hanging in space and not on the steering wheel
  • Anatomy is slightly distorted with human-like legs emerging from the jacket

Vidu Q2

  • + Perfectly adhered to the complex prompt by including both the driving capybara and the bored passenger
  • + Successfully captured the requested pose with both paws on the steering wheel
  • + Composition feels more like a complete scene rather than a portrait
  • The passenger's hair and face have some slight blurring/artifacts
  • The capybara's hands look more like primate hands than capybara paws

Verdict: Stable Diffusion 3.5 Large failed on a critical part of the prompt by omitting the passenger entirely, resulting in a high-quality but incomplete image. Vidu Q2 successfully followed all instructions, including the passenger, the specific poses, and the bored expression, making it the clear winner for prompt adherence despite slightly lower fine detail in the background.

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.”

Stable Diffusion 3.5 Large
Vidu Q2

AI Judge Analysis

Stable Diffusion 3.5 Large

  • + Strong cinematic lighting and atmosphere with the large glowing moon.
  • + Highly detailed parchment texture with realistic torn edges.
  • + Mostly accurate rendering of requested spooky elements like twisted trees and gravestones.
  • Completely failed to include the event details (Date, Time, Location) at the bottom.
  • The text on the scroll banner becomes illegible and garbled at the end.
  • The jack-o-lantern is off-center rather than being the 'central' element.

Vidu Q2

  • + Excellent adherence to the 'central jack-o-lantern' instruction.
  • + Successfully included all event details at the bottom of the poster.
  • + Accurately rendered the thorny border and webs as requested.
  • Frequent spelling errors in both the title ('Intovztion') and the banner ('invieed', 'fiigts').
  • Wrong date used (2025 instead of 2026) and '7pm' is rendered as 'Tmm'.
  • The composition feels a bit more like a digital collage than a cohesive vintage poster.

Verdict: Stable Diffusion 3.5 Large creates a much more atmospheric and visually impressive 'vintage' piece with superior lighting, but it failed significantly on prompt adherence by omitting the specific event details. Vidu Q2 followed the prompt's layout and content instructions much better, including all required text fields, but suffered from significant spelling errors and a slightly less 'polished' artistic style. Vidu Q2 is the marginal winner for capturing all specific text requirements, despite the typos.

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.”

Stable Diffusion 3.5 Large
Vidu Q2

AI Judge Analysis

Stable Diffusion 3.5 Large

  • + Highly detailed textures on the fish and rice grains
  • + Excellent lighting and realistic rendering of the base materials
  • + Correct text spelling and inclusion of requested items like chopsticks
  • The text is placed on a sign rather than being floating/centered as requested
  • The composition feels slightly cluttered rather than minimal

Vidu Q2

  • + Fits the '3D cartoon' aesthetic much better than the other model
  • + Perfectly adheres to the text placement and layout instructions
  • + Very clean composition with a clear diorama feel
  • The texture of the rice is less realistic and more plasticky
  • Lower level of fine detail in the sushi components

Verdict: Stable Diffusion 3.5 Large creates a much more realistic and detailed image with impressive PBR textures, but it misses the specific layout instructions regarding floating text. Vidu Q2 perfectly follows the composition and text placement prompts and captures the 'cartoon miniature' style more accurately, making it the more successful image for this specific prompt.

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.”

Stable Diffusion 3.5 Large
Vidu Q2

AI Judge Analysis

Stable Diffusion 3.5 Large

  • + Excellent soft lighting with a strong morning dew aesthetic
  • + High fur detail and naturalistic textures
  • + Great sense of movement and dynamic action
  • Anatomical issues with the kitten, which lacks proper paws and has a deformed body
  • The kitten does not follow the 'tabby' description, appearing more like an orange-ginger mix
  • Butterfly on the rabbit's head is merged awkwardly

Vidu Q2

  • + Successfully includes all animals with accurate markings, including a distinct tabby kitten
  • + Better composition showing the full environment and playful interaction
  • + Stronger adherence to the 'god rays' lighting request
  • The fox kit has a slightly cartoonish face and an anatomically awkward front leg
  • The second golden retriever puppy was not requested in the prompt
  • Less focus on the 'macro' ultra-detailed fur texture compared to Model A

Verdict: Stable Diffusion 3.5 Large offers superior texture and cinematic lighting, but suffers from significant anatomical distortion in the kitten. Vidu Q2 follows the prompt much more accurately by including all four distinct animals with their specific requested markings (like the tabby stripes), and despite adding an extra puppy, provides a more coherent and joyful scene.

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.”

Stable Diffusion 3.5 Large
Vidu Q2

AI Judge Analysis

Stable Diffusion 3.5 Large

  • + Excellent typography rendering with almost perfect spelling
  • + Clean vector emblem style with a professional layout
  • + Strong adherence to the color palette and vintage texture request
  • Spelled 'Caffè' with an extra 'e' as 'Cafféé'
  • The steam effect is a bit chunky and less refined than the rest of the logo

Vidu Q2

  • + Nice interpretation of the cloche dome with elegant steam
  • + Good use of warm brown and cream tones
  • Poor typography with significant spelling errors ('Farmiin', 'Esttt', 'Fopli20')
  • Text layout is cluttered and repetitive
  • Fails to present a clean vector emblem style

Verdict: Stable Diffusion 3.5 Large is the clear winner as it produces a professional, usable logo with clean vector lines and nearly perfect text. Vidu Q2 fails significantly on the typography and overall composition, resulting in a cluttered image with multiple spelling hallucinations.

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.”

Stable Diffusion 3.5 Large
Vidu Q2

AI Judge Analysis

Stable Diffusion 3.5 Large

  • + Features a sophisticated, complex layout that looks like a technical blueprint.
  • + Adheres closely to the requested navy, white, and muted red color palette.
  • + Includes high-quality textures on the Earth and Moon surfaces.
  • Shows a Space Shuttle instead of the Saturn V rocket requested for the Apollo mission.
  • The layout is cluttered and does not follow the requested step-by-step logic.
  • Text is completely illegible illegible gibberish.

Vidu Q2

  • + Excellent adherence to the 'flat-vector' and 'infographic' style with clean icons.
  • + Follows the sequential structure of the mission steps much more clearly.
  • + Provides much better text legibility, even if several words are misspelled.
  • Failed to use a Saturn V rocket icon, showing a generic small rocket instead.
  • The overall color palette is a bit washed out and lacks the deep navy requested.

Verdict: Stable Diffusion 3.5 Large produced a visually rich technical poster but failed significantly on the specific content instructions, including the wrong spacecraft (a Shuttle) and a total lack of step-by-step logic. Vidu Q2 followed the infographic structure perfectly, providing clean, consistent iconography that clearly maps to the requested mission phases, despite some spelling errors. Vidu Q2 is the preferred choice for actually meeting the infographic design brief.

Next steps

Explore each model