Vidu Q2 ShengShu Technology vs Wan 2.7 Alibaba

Vidu Q2

• + Excellent adherence to the glass cube prompt with realistic thick edges.
• + Very high visual quality and realistic textures on the book and sphere.
• + Beautiful lighting and shadow casting that matches the described window light.

• − The plant is more overhead than 'behind' the cube, though still partially visible through it.
• − The glass cube lacks a physical top panel, allowing the book to seemingly float or rest on the edges.

Wan 2.7

• + Accurately places the plant behind the cube as requested.
• + Correctly renders reflected light and internal reflections within the glass.
• + Good photorealistic texture on the weathered wooden table.

• − The internal geometry of the glass cube is inconsistent, with some edges appearing like mirrors.
• − The blue sphere has a slightly rough, mottled texture compared to the smooth sphere in the other image.

Vidu Q2

• + Excellent preservation of the car's design and details.
• + The man from the source image is clearly recognizable and logically integrated into the driver's seat.
• + High visual quality with realistic motion blur and accurate lighting.

• − The man's scale relative to the car is slightly small.
• − Minor perspective issue where the car seems to be angled slightly differently than the road lane.

Wan 2.7

• + Natural lighting that blends the car into the coastal environment.
• + Good composition with the winding road and coastline.

• − Completely failed to use the man from the source image, substituting a distorted generic person.
• − The driver's face is severely malformed.
• − Lower resolution and more painterly texture compared to the source.

Vidu Q2

• + Excellent skin texture and facial details
• + Captured the 'imperfect framing' prompt with a tight, candid crop
• + Very convincing wet textures and refractions

• − Anatomical errors in the hands and several impossible bicycle parts
• − Background car is static with no motion blur as requested

Wan 2.7

• + Strong environment storytelling with a believable Japanese street scene
• + Captures the full silhouette and action of the man
• + Good composition and wet pavement reflections

• − Lacks motion blur on passing cars
• − Lower resolution/clarity on the subject's face compared to Model A
• − The street feels a bit 'clean' for a candid/imperfect frame prompt

Vidu Q2

• + Excellent metallic texture with realistic wear and orange-toned lighting reflections.
• + Beautiful intricate engraving on the armor and rich texture on the leather straps.
• + Superior cinematic lighting that creates more depth and visual interest.

• − The facial features look slightly too clean or 'airbrushed' for a battle-worn character.
• − Lower density of braids compared to the prompt's likely intent.

Wan 2.7

• + Perfect adherence to the 'hair braided with small beads' requirement with multiple visible braids.
• + Excellent skin texture featuring more realistic battle damage, dirt, and age lines.
• + Very strong adherence to the 'lifelike eyes' and 'faint scars' descriptors.

• − The armor engravings are slightly less sharp and detailed compared to the leather and skin.
• − The torch in the background is a bit distracting and less 'bokeh' than the abstract sparks in Model A.

Vidu Q2

• + Features multiple vibrant food photos with good clarity
• + Includes sections for appetizers, pizza, and mains as requested

• − Text consists of gibberish characters and severe letter distortions
• − Graphic elements like columns overlap and feel cluttered

Wan 2.7

• + Excellent typography with legible, realistic English text
• + Perfectly organized grid layout that follows the minimalist prompt
• + Professional composition featuring realistic restaurant branding and contact info

• − The layout is more of a flyer or digital menu than a traditional multi-page menu
• − Background includes out-of-frame props like a pen and herbs which weren't specifically requested

Vidu Q2

• + Excellent photorealistic texture on the meat patty and buns
• + Strong, vibrant lighting that blends the subject into the fiery background
• + Clean and legible primary typography

• − The currency symbol is garbled/incorrect in the price tag
• − The burger arrangement is more 'stacked' than 'exploded' compared to the other model

Wan 2.7

• + Accurate rendering of the Euro currency symbol (€)
• + Dynamic exploded composition with more separation between ingredients
• + Consistent fiery effect applied across all text elements

• − The food textures look slightly more illustrative/rendered than photorealistic
• − Random flying seeds and cucumber slices were not explicitly requested

Vidu Q2

• + Excellent realistic chalk texture with smudges and natural variations
• + Handwriting feels organic and consistent with a hand-drawn style

• − Numerous spelling errors including 'Musshoom', 'Lemepun', and 'Octopd'
• − Incorrect price on the first item and garbled text at the bottom

Wan 2.7

• + Perfect text accuracy and spelling for all menu items
• + Clean and legible layout that follows all prompt instructions
• + Correct pricing and dates as requested

• − The text looks like a digital font rather than authentic hand-lettered chalk
• − Lacks the organic smudging and 'dusty' realism of a real chalkboard

Vidu Q2

• + Successfully integrated nearly all character details from Image 2, including the sunglasses, scarf, and black sweatshirt.
• + Accurately replicated the complex leg-crossing pose and overall body position from Image 1.
• + Matched the yellow background and red ottoman from the source environment.

• − The hands have anatomical errors, specifically the lower left hand which has too many fingers.
• − The character's expression is slightly altered compared to the neutral confidence seen in Image 2.

Wan 2.7

• + Preserved the original composition of Image 1 perfectly.

• − Completely failed to perform the edit, essentially returning the source Image 1 with minor color adjustments.
• − Did not incorporate the character, clothing, or attributes from Image 2 at all.

Vidu Q2

• + Vibrant color palette with nebula-like skin on the horse enhancing the surreal theme.
• + High level of cinematic lighting and glowing particles.
• + Good integration of the character into the fantastical environment.

• − The astronaut's left leg position is anatomically awkward relative to the horse's back.

Wan 2.7

• + Excellent anatomical rendering of both the astronaut's suit and the horse.
• + Clearer background elements with realistic planets and galaxies.
• + Better composition and use of negative space.

• − The horse's design is slightly conventional and less 'surreal' than the horse in Vidu Q2.

Vidu Q2

• + Successfully transferred the exact clothing items (peacoat, plaid scarf, watch, sunglasses) from Image 2.
• + Maintains the skin vitiligo patterns on the arm with high consistency to Image 1.
• + Accurately places the sand texture on the arm and face.

Wan 2.7

• + Maintains the subject's exact facial features and skin condition significantly better than the competitor.
• + Preserves the original background and composition of Image 1 perfectly.
• + Naturally integrates the vitiligo pattern onto the hands and neck.

• − Completely failed to use the outfit from Image 2, generating a generic black and gold suit instead.
• − Incorrectly changed the pose of the subject's legs from Image 1.
• − Added excessive jewelry that was not present in either source image.

Vidu Q2

• + Excellent internal space logic with the passenger correctly placed in the rear seat back.
• + High-quality lighting and reflections on the car window and interior surfaces.
• + Detailed fur texture and realistic interaction between the capybara's paws and the steering wheel.

• − The passenger's phone is slightly distorted/elongated.
• − The driver's jacket looks more like a modern mechanic's shirt than a traditional uniform.

Wan 2.7

• + The capybara's expression is very calm and fits the prompt well.
• + The taxi rooftop light is visible, adding to the exterior realism.

• − Failed the spatial prompt by putting the passenger in the front seat instead of the back.
• − The capybara's paws are rendered with strange, multiple sharp claws that look unnatural.
• − The car's door pillar is missing, creating a nonsensical joined window space.

Vidu Q2

• + Features more cinematic lighting with a glowing parchment effect.
• + Dynamic border with thorns and cobwebs that feels more organic.

• − Several severe spelling errors including 'Intovigtion' and 'invieed'.
• − Incorrect date and time formatting (30.70.2025 and Tmm).

Wan 2.7

• + Perfect text rendering for all requested fields including the title and event details.
• + Highly detailed and clean vintage illustration style with consistent borders.
• + Accurately follows the specific date (30.10.2026) and instructions.

• − The 'scroll banner' is more of a flat ribbon at the bottom rather than a central feature.
• − Slightly less 'moody' or dark than the prompt suggested, leaning towards a clean illustration.

Vidu Q2

• + Successfully adds thick hair while keeping the background and clothing identical.
• + The hair texture matches the rugged aesthetic of the original image.

• − The hairline is a bit messy and slightly obscures the glasses frame.
• − The forehead area shows some minor blending artifacts where the hair meets the skin.

Wan 2.7

• + Adds a very natural, well-integrated head of hair with realistic volume.
• + Preserves facial features, glasses, and lighting perfectly.
• + Excellent blending at the temples and hairline.

• − The hair is slightly less 'thick' than Model A, though still very realistic.

Vidu Q2

• + Excellent typography with a clean, pillowy 3D effect.
• + High-quality material rendering for the fish and rice textures.
• + Creative interpretation of the flag as an physical icon within the scene.

• − Lighting is a bit harsh on the sushi toppings leading to blown-out highlights.
• − The base diorama has strange, unidentifiable blobs on the corners.

Wan 2.7

• + Perfect adherence to the 45-degree isometric perspective.
• + Very clean and organized composition that emphasizes the 'miniature' feel.
• + Accurate rendering of multiple sushi types and condiments like wasabi and ginger.

• − The text 'JAPAN' is slightly misaligned horizontally.
• − The flag icon is a simple 2D graphic rather than a 3D asset.

Vidu Q2

• + Excellent caricature style that maintains a strong likeness to the source person.
• + Successfully integrates hockey (rink background, puck, net), TV news (microphone, desk), and dogs in a cohesive scene.
• + High visual clarity with clean lines and vibrant colors.

• − The fingers on the hand holding the microphone are anatomicaly distorted.
• − The small dog on the right is standing on a stack of papers in a somewhat awkward way.

Wan 2.7

• + Creative use of speech bubbles to emphasize the themes of the request.
• + Includes a humorous detail of a dog wearing a hockey helmet.
• + Preserves the selfie-style pose from the source image.

• − The facial features are significantly altered, losing much of the likeness from the source photo.
• − Contains a spelling error in the speech bubble ('Rolee' instead of 'Rule').
• − The composition feels a bit cluttered with overlapping elements.

Vidu Q2

• + Excellent dynamic composition with a sense of playful movement.
• + Beautiful color palette with vibrant wildflowers and warm lighting.
• + Very soft, detailed fur textures on all animals.

• − Includes two golden retriever puppies instead of one.
• − The fox kit has slightly stylized, almost cartoonish eyes compared to the others.

Wan 2.7

• + Perfect adherence to the count of animals requested.
• + Great depiction of 'god rays' and dew sparkles as requested in the prompt.
• + High-quality fur detail and naturalistic lighting on the subjects.

• − The kitten's pose and anatomy (standing on the bunny) look a bit awkward.
• − The composition is more static and cluttered compared to the other image.

Vidu Q2

• + Excellent adherence to the Studio Ghibli art style including eye and mouth shapes.
• + Maintains clear, vibrant, and clean line work typical of modern anime.
• + Perfectly captures the exaggerated expressions requested in an illustrative format.

• − The man's hand on his hip looks a bit malformed compared to the source.

Wan 2.7

• + Successfully applies a soft, hand-painted watercolor texture.
• + Preserves the facial features and likeness of the original people more accurately.
• + Excellent use of warm, nostalgic, and dreamy lighting.

• − Does not fully commit to the specific 'Ghibli' aesthetic, leaning more toward a general watercolor portrait.
• − The foreground woman's face looks slightly uncanny between realism and illustration.

Vidu Q2

• + Excellent adherence to the 'flying leaves' request with a large quantity of colorful leaves.
• + Effective hair motion that creates a clear sense of wind direction.
• + Perfect preservation of the face, clothing, and background details.

Wan 2.7

• + Natural and realistic hair flow that feels integrated with the subject.
• + Strong preservation of the original image's lighting and colors.
• + Good sense of motion without overwhelming the scene.

• − The 'flying leaves' are significantly less prominent compared to the result in Image A.
• − Several leaves appear as blurry brown smudges rather than distinct leaf shapes.

Vidu Q2

• + Successfully captures the requested steam effect inside the cloche.
• + Good use of color tones matching the warm brown and cream request.

• − Significant spelling errors throughout the logo including 'Farmiin' and 'Fopli20'.
• − The typography is messy and overlaps with the banner border.

Wan 2.7

• + Excellent layout with a clean vector emblem style.
• + Very accurate spelling for the 'Est. 1720' banner and legible typography.
• + Excellent source preservation and subtle paper texture on the background.

• − Mistyped 'Florian' as 'Florion'.
• − The steam effect is a bit static and illustrative rather than atmospheric.

Vidu Q2

• + Clean vector illustration style.
• + Distinct and high-quality iconography for the lunar modules.

• − Nonsense text for almost every label.
• − Logical flow is disjointed and does not follow the requested 1-6 step sequence.
• − Palette is overly desaturated compared to the NASA-inspired request.

Wan 2.7

• + Excellent text rendering with legible and accurate labels.
• + Follows the requested 6-step mission sequence perfectly.
• + High adherence to the NASA-inspired color palette and modern infographic layout.

• − Small minor typo in 'DESCRIPT' (meant to be DESCENT).
• − Minor typo in 'Tranquiliry'.