Wan 2.6 Alibaba vs Wan 2.7 Alibaba

Wan 2.6

• + Excellent depiction of soft window lighting and realistic shadows.
• + Beautiful internal reflections and refractions through the glass cube.
• + High visual quality with a pleasing bokeh effect on the plant.

• − The glass cube has extra vertical internal edges that make it look like multiple panes rather than a simple cube.

Wan 2.7

• + Realistic textures on the wooden table and the book's spine.
• + Good placement of the plant behind the glass cube.
• + Clean geometry for the glass cube's exterior.

• − The blue sphere reflection on the left is physically incorrect as if it were a mirror, not glass.
• − The sphere has a slightly rough, non-glass texture compared to Model A.

Wan 2.6

• + Excellent preservation of the man's identity, including his specific hairstyle, scarf, and coat pattern.
• + High-quality action scene with realistic motion blur in the wheels and background.
• + Accurate car model preservation based on the source image.

• − The man is seated on the wrong side for a standard US road (UK/LHD configuration), though he is steering.

Wan 2.7

• + Successfully places the car on a coastal California road with golden hour lighting.
• + Maintains the car's exterior design and color well.

• − Completely fails to preserve the identity of the man, replacing him with a distorted, unrecognizable figure.
• − The person in the driver's seat is low-resolution and lacks any resemblance to the source image provided.

Wan 2.6

• + Excellent shallow depth of field and bokeh that matches a 50mm lens profile.
• + Superb skin texture and life-like details on the man's hands and face.
• + Perfect execution of the wet pavement reflections and rainy atmosphere.

• − The raindrops on the man's jacket look a bit like static beads or glitter rather than soaked fabric.

Wan 2.7

• + Naturalistic 'candid' composition and wider framing of a Japanese street.
• + Realistic clothing textures showing dampness rather than just surface droplets.

• − Failed to incorporate the 'motion blur from passing cars' requested in the prompt.
• − The mechanical structure of the bicycle becomes nonsensical near the pedals and chain area.
• − Lacks the shallow depth of field requested, with much of the background remaining relatively sharp.

Wan 2.6

• + Excellent representation of ornate engraved plate armor with high-contrast reflections.
• + Superior lighting and atmosphere with vivid torchlight and realistic skin textures.
• + Very detailed treatment of the hair beads and fraying cloth underlayer.

• − The dirt on the face looks slightly like flat texture overlays in some spots.
• − The composition is very tightly cropped on the left side.

Wan 2.7

• + Strong symmetry and clear view of the complex braided hairstyle.
• + Clean, sharp rendering of facial features and scars.
• + Good depth of field with distinct bokeh elements in the background.

• − The armor engraving is less intricate and appears flatter than Model A.
• − Lighting feels a bit more generic and less like actual warm torchlight reflecting off the metal.

Wan 2.6

• + Successfully uses a grid layout as requested
• + Includes specific sections for Appetizers, Pizza, and Mains
• + Features bold sans-serif fonts in a clean, professional manner

• − The text is mostly gibberish or misspelled ('RESTAUR MENU')
• − The grid lacks food variety, showing mostly pizza and salad
• − The layout feels slightly more like a flyer than a functional menu

Wan 2.7

• + Excellent text legibility and coherent menu items
• + High food variety in the grid, including pasta, burgers, and desserts
• + Sophisticated composition including lifestyle props like a pen and herbs

• − The food photos have rounded corners instead of the sharp minimalist grid often associated with the prompt
• − Does not follow the specific 'Mains' text section list, instead putting prices under photos

Wan 2.6

• + Excellent photorealistic texture on the burger bun and patty.
• + Seamless integration of the fiery background with realistic smoke and lighting.
• + High-quality rendering of the main titles with a convincing flaming effect.

• − The price starburst looks like a flat 2D sticker compared to the 3D scene.
• − A sauce stream appears to be unrealistically originating from inside the top bun.

Wan 2.7

• + Stronger 'exploded' composition with more dynamic spacing of the ingredients.
• + Price starburst is better integrated into the overall fiery theme and lighting.
• + Clean, professional typography that is very easy to read.

• − The food items look more like 3D digital illustrations than photorealistic photography.
• − Inclusion of random floating seeds/nuts that weren't requested and don't fit the burger theme.

Wan 2.6

• + Excellent chalk texture with realistic smudges and dusty residue
• + Truly handwritten appearance with natural variations in letter size and slant
• + Perfect text accuracy including the clipped prompt text completion

• − The 'cursive' requirement for the title is only partially met with print/cursive hybrid letters

Wan 2.7

• + Perfect text rendering without any spelling errors
• + Clean and centered composition
• + Attractive cafe-style background lighting

• − Text looks like a digital font rather than natural chalk handwriting
• − Fails the 'no printed or digital fonts' requirement
• − Lacks the specific chalky texture and variations requested in the prompt

Wan 2.6

• + Excellent cinematic lighting and atmospheric effects
• + High level of detail in the astronaut suit and horse's mane

• − Completely failed the negative constraint to put the horse on top of the astronaut
• − The horse's back right leg is missing its lower half

Wan 2.7

• + Clear and coherent composition with sharp focus
• + Good rendering of the horse's anatomy and planets

• − Completely failed the negative constraint to put the horse on top of the astronaut
• − Lacks the 'surreal' and 'cinematic' lighting depth requested

Wan 2.6

• + Successfully transferred the pea coat, scarf, and sunglasses from Image 2.
• + Maintained the perspective, lighting, and placement of the person in relation to the wooden structure.
• + Preserved the sand patterns on the face and the hair detail accurately.

• − Cropped the image significantly compared to the original aspect ratio.
• − Failed to include the lower half of the outfit (jeans, belt) due to the tight crop.
• − The scarf pattern, though similar, is not an exact match to Image 2.

Wan 2.7

• + Successfully integrated a full-body outfit, including shoes, into the scene.
• + Maintained the full original composition and background elements well.
• + Handled the vitiligo patterns on the hands realistically beneath the new clothing.

• − Completely ignored the clothing in Image 2, generating a generic gold-embroidered coat instead.
• − Failed to include the sunglasses and scarf depicted in the source image.
• − Changed the person's pose slightly, making him stand away from the pillar rather than leaning against it.

Wan 2.6

• + Excellent photorealism with realistic low-light noise and rain textures
• + Captures the 'bored' expression of the passenger perfectly
• + The capybara's pose and anatomy look more integrated with the driver seat

• − The capybara's paws are somewhat indistinct and blend into the steering wheel

Wan 2.7

• + Clearer depiction of the capybara's paws on the steering wheel
• + Very sharp image resolution and clean lighting

• − The capybara's fur has a slightly artificial, 'rendered' look compared to Model A
• − The passenger is positioned awkwardly close to the driver, making the backseat feel like a front seat

Wan 2.6

• + Excellent atmospheric cinematic lighting with a realistic 3D feel.
• + Highly stylized texture on the main title text that fits the gothic theme.
• + The border design perfectly integrates webs and thorns as requested.

• − The parchment texture is less 'poster-like' and more of an overlay effect.
• − Small amount of background text is slightly less crisp than the main header.

Wan 2.7

• + Perfect text rendering for all lines including the specifically requested event details.
• + Clean and detailed vintage illustration style with more thematic elements like ravens and a cauldron.
• + Excellent parchment aesthetics that feel like a physical printed invitation.

• − The art style is more like a 2D illustration rather than the 'cinematic lighting' requested.
• − The 'webs and thorns' border is a bit more repetitive and flat compared to Model A.

Wan 2.6

• + Successfully added a full head of hair that matches the requested description.
• + Maintains high resolution and realistic hair texture.

• − Noticeably alters the subject's face, making him look like a younger/different person.
• − The hairline integration with the forehead looks slightly forced compared to the original bone structure.

Wan 2.7

• + Excellent preservation of the original facial features and identity.
• + The hairline and hair growth pattern look incredibly natural and well-integrated with the scalp.
• + Near-perfect source preservation for all elements outside the hair area.

• − Small minor artifacting where the hair meets the side of the glasses frame.

Wan 2.6

• + Excellent typography rendering and placement
• + Very clean, minimal aesthetic that fits the 'miniature' prompt perfectly
• + Accurate 45-degree isometric perspective

• − The diorama base is a bit plain and simple
• − Includes flag icon on the same line as text instead of below

Wan 2.7

• + High level of detail in the food textures and materials
• + More variety in the sushi pieces while maintaining a clean look
• + Sophisticated rendering of the diorama base and dish elements

• − The text has slight visual artifacts (dark outlines) compared to the clean white of Model A
• − The text layout is slightly off-center

Wan 2.6

• + Successfully incorporates all prompt elements (TV anchor, dogs, hockey) in a balanced scene.
• + Good caricature style that retains reasonable likeness to the source person.
• + The TV studio setting is clear and professional-looking for a caricature.

• − The hockey stick has an awkward, extra-long handle that disappears behind her head.
• − The facial expression is a bit generic and loses some of the source subject's unique features.

Wan 2.7

• + Very creative and humorous interpretation with dogs wearing helmets and a miniature hockey rink.
• + Excellent preservation of the original selfie-style composition and high likeness to the source subject.
• + Text rendering in speech bubbles is clean and legible.

• − One of the speech bubbles contains a typo ('Dogs Rolee!' instead of 'Rule').
• − The 'TV anchor' profession is represented only by a small lapel mic and a screen, making it less obvious than model A.

Wan 2.6

• + Exceptional god ray effects and atmospheric lighting
• + Highly dynamic poses that feel like they are 'tumbling' as requested
• + Detailed fur textures and realistic interaction between the animals

• − The fox's anatomy in the bottom right is slightly distorted during the tumble

Wan 2.7

• + Clearer distinctiveness of the different flower types
• + Good rendering of the golden retriever's face and fur
• + Accurate inclusion of all four requested animal types

• − The composition feels more static and posed rather than playful/tumbling
• − The kitten has an oddly protruding tongue/mouth artifact
• − God rays are less realistic and more like graphic lines

Wan 2.6

• + Excellent character translation into a classic anime aesthetic.
• + Consistent watercolor texture throughout the image.
• + Dreamy lighting and soft pastel palette perfectly match the prompt.

• − Added sparkles/dust motes might feel a bit distracting to some users.

Wan 2.7

• + Strong preservation of the source image's facial features.
• + Good hand-painted texture and soft color palette.
• + Excellent background coherence.

• − The faces lean more toward 'digital painting' than true Ghibli-style illustration.
• − Slightly less 'dreamy' lighting compared to the alternative.

Wan 2.6

• + Excellent wind-blown hair effect that looks natural and fluid
• + Vibrant green leaves added that match the park environment
• + High fidelity preservation of the original woman and dog

• − Leaves look a bit static and lack motion blur

Wan 2.7

• + Added a high volume of flying leaves to create a sense of motion
• + Brown leaves suggest an autumn breeze effect
• + Good preservation of the source image identity

• − The hair effect is slightly less cohesive than Model A
• − The lighting on the brown leaves doesn't perfectly match the sunny summer day lighting of the scene

Wan 2.6

• + Perfectly rendered text for both the name and the banner
• + Clean minimalist vector execution with effective use of negative space
• + Accurate representation of an 'Est. 1720' banner integrated into the emblem

• − The cloche shape is a bit thick, resembling a bowl or lid more than a traditional cloche

Wan 2.7

• + Excellent vintage badge composition with a sophisticated color palette
• + High-quality subtle texture on the background and within the emblem
• + Refined line art for the cloche and steam elements

• − Typos in the main text, spelling it 'Florion' instead of 'Florian'
• − The design is quite busy and borders on decorative rather than 'minimalist' as requested

Wan 2.6

• + Legible main title
• + Accurately identifies the three crew names

• − Fails to include any of the requested 6 stages of the infographic
• − Composition is empty and lacks icons
• − The background texture looks like a printed towel or fabric rather than a vector poster

Wan 2.7

• + Perfectly follows all 6 requested steps with relevant icons
• + Excellent adherence to the NASA-inspired color palette and vector style
• + High-quality typography and information density
• + Strong visual composition following a logical flow

• − Minor spelling errors in smaller text like 'Descript' and 'Tranquiliry'
• − The map pin logic for 'Tranquiliry' is slightly disconnected from the main flow line