Grok Imagine Image Pro xAI vs Vidu Q2 ShengShu Technology

Grok Imagine Image Pro

• + Excellent realism in the texture of the wooden table and the glass pane reflections.
• + Highly accurate depiction of light refraction through the glass cube.
• + Legible and context-relevant text on the book spine.

• − The plant is more 'behind' the cube than 'visible through' it, as some leaves overlap the top.

Vidu Q2

• + Vibrant colors and strong lighting contrast that follows the 'light from left' instruction.
• + Good composition with interesting shadows falling across the table.

• − Physics issues with the reflection of the blue sphere which looks more like a second sphere underneath.
• − The glass cube edges appear slightly distorted and inconsistent in thickness.
• − The book is floating slightly above the glass surface.

Grok Imagine Image Pro

• + Excellent preservation of the car's model and design features.
• + High visual quality with a realistic sense of motion on the road.
• + Beautifully composed California coastline background.

• − Completely failed to use the specific man from the source image, replacing him with a different person.
• − The steering wheel position seems slightly off compared to the driver's grip.

Vidu Q2

• + Excellent subject preservation, correctly identifying and placing the man from the source image behind the wheel.
• + Good preservation of the source car's identifying features.
• + Dynamic composition with a sense of speed.

• − The man's hands on the steering wheel have anatomical errors (extra/deformed fingers).
• − The car's proportions feel slightly elongated compared to the source.

Grok Imagine Image Pro

• + Excellent full-body composition and anatomy
• + Realistic lighting and reflections that match the rainy environment
• + Strong adherence to the 'candid' and 'motion blur' instructions

• − The wrench he is holding doesn't align correctly with the bike's nut
• − Background storefronts look slightly generic

Vidu Q2

• + Successfully captures a very shallow depth of field
• + Grit and realistic texture on the bicycle frame and man's skin

• − Major anatomical errors with three hands visible
• − Strange bicycle geometry with a second wheel appearing in the foreground
• − Failed to include motion blur for passing cars

Grok Imagine Image Pro

• + Excellent adherence to the hair braiding and beads requirement, with detailed placement throughout the hair.
• + Superior rendering of textures, particularly in the weathered leather, rusted engravings, and cloth underlayer.
• + Very convincing skin textures with realistic scars, dirt, and lifelike eye detail.

• − The placement of beads on the top of the head appears slightly artificial or stuck on.
• − The bokeh sparks are a bit uniform in size across the frame.

Vidu Q2

• + Strong dynamic lighting with warm torchlight glints on the edges of the armor plates.
• + Composition feels slightly more natural with a three-quarter view rather than a direct frontal shot.
• + Good depiction of battle damage including blood spatter and dents on the metal.

• − Minimal braiding and beads compared to the prompt's specific request.
• − Texture on the cloth underlayer is less defined and detailed than the competing model.
• − Armor engravings appear slightly more generic and less 'ornate'.

Grok Imagine Image Pro

• + Excellent grid layout that strictly follows the requested sections for Appetizers, Pizza, and Mains.
• + High-quality, realistic food photography that looks professional and appetizing.
• + Mostly legible text including prices and descriptions that make sense in a menu context.

• − The placeholder text description for 'Avocado Toast' incorrectly repeats the description for 'Prosciutto Arugula'.
• − Uses a slightly stylized, handwritten-looking font for descriptions instead of a strictly bold sans-serif throughout.

Vidu Q2

• + Successfully incorporates vibrant accents and a modern minimalist aesthetic.
• + Good distribution of colorful food images across the page.

• − Text is largely gibberish or contains severe misspellings (e.g., 'APECIZEN', 'MPAIZERS').
• − Layout is chaotic and does not follow a clear, professional logical flow like a real menu.
• − Food images are repetitive and some contain visual artifacts/distortions.

Grok Imagine Image Pro

• + Excellent photorealistic texture on the meat patty and fresh vegetables.
• + Dynamic 'exploded' composition with unique cheese-pull effects between layers.
• + Highly accurate text rendering, including the specific Euro currency symbol and price format.

• − The starburst for the price is a bit generic and flat compared to the rest of the image.
• − Background feels slightly more like floating embers than a full 'fiery' environment.

Vidu Q2

• + Stronger 'fiery' background that fills the frame with intense energy.
• + Good glow effect applied to all text layers as requested.
• + Vibrant colors that pop against the dark backdrop.

• − The currency symbol is malformed and does not resemble a Euro symbol.
• − The explosion of ingredients feels less layered and physics-accurate than Image A.
• − The bottom bun texture appears somewhat artificial.

Grok Imagine Image Pro

• + Perfect text accuracy and spelling for all menu items.
• + Authentic chalk texture and realistic smudge marks on the board.
• + Clean composition that remains legible and follows the prompt's formatting perfectly.

• − The handwriting is a bit uniform, arguably leaning toward a 'chalk font' aesthetic rather than purely messy handwriting.

Vidu Q2

• + Very expressive and varied handwriting style with thick chalk strokes.
• + Good use of color variance in the chalk highlights.

• − Significant spelling errors throughout the text (e.g., 'Musshoom', 'Lemepun', 'Browd Botter').
• − Changed the price of the first item from $24 to $34.
• − The text becomes garbled and illegible at the bottom of the board.

Grok Imagine Image Pro

• + Successfully maintains the yellow background and red ottoman from the source image.

• − Completely failed the character instruction by simply re-rendering the original woman.
• − Did not incorporate any clothing elements, face, or features from Image 2.

Vidu Q2

• + Excellent character transfers, including the face, sunglasses, scarf, and black sweatshirt.
• + Perfectly replicates the complex leg-cross and arm-extension pose from Image 1.
• + Accurately matches the lighting of the yellow studio environment from Image 1.

• − Hand anatomy on the lower left arm has too many fingers.
• − The scarf physics are slightly unnatural given the lean of the character.

Grok Imagine Image Pro

• + Successfully followed the specific instruction of having the horse on top of the astronaut
• + Highly detailed rendering of both the spacesuit and the horse's musculature
• + Strong cinematic composition with the planet and nebula backgrounds

• − The horse is floating just above the astronaut rather than literally 'riding' him
• − Anatomical weirdness where the horse's back leg appears to be emerging from its front or belly

Vidu Q2

• + High visual quality with vibrant colors and cosmic patterns integrated into the horse's coat
• + Excellent clarity and lighting coordination between the subjects and the environment

• − Failed the negative constraint entirely by placing the astronaut on top
• − Lacks the surreal 'subversion' requested in the prompt

Grok Imagine Image Pro

• + Excellent preservation of the subject's facial features and specific vitiligo patterns on the forehead.
• + High visual quality and realistic fabric rendering.
• + Keeps the background architectural elements and lighting consistent with Image 1.

• − Failed completely to use the outfit from Image 2, substituting it with a generic royal/fantasy costume.
• − The hands are significantly lightened and do not match the subject's actual skin tone.

Vidu Q2

• + Successfully transferred the specific outfit from Image 2, including the coat, scarf, sunglasses, and watch.
• + Accurately adapted the clothing to the subject's pose and environment.
• + Maintains the subject's skin condition and overall likeness.

• − The glasses and mustache from Image 2 were blended into the subject's face, slightly altering the original face preservation.
• − The hand in the bottom left has anatomical distortions (too many fingers).
• − Added sand textures to the coat that were not present in the original outfit.

Grok Imagine Image Pro

• + Excellent text rendering on the driver cap mentioning 'NYC TLC Medallion'.
• + Superior photorealistic lighting and interior detail.
• + The businesswoman is correctly positioned in the back seat.

• − The passenger is sitting in the middle/right rather than fully behind the driver as a taxi usually operates, though it still works.
• − The capybara's paws look slightly claw-like and less like natural paws.

Vidu Q2

• + Natural side-profile composition showing the depth of the car.
• + Good adherence to the requested calm expression on the capybara.
• + The businesswoman captures the 'bored' expression very effectively.

• − The businesswoman is placed in the seat directly behind the driver but appears physically separated or lacks clear spatial coherence at the bottom of the frame.
• − The capybara paws look like human hands covered in fur.

Grok Imagine Image Pro

• + Excellent typography with perfect spelling in all requested text fields.
• + Superb atmospheric lighting and atmospheric depth in the illustration.
• + High degree of adherence to the composition requirements, including the scroll banner and parchment texture.

• − The scroll banner placement is a bit central rather than integrated into the top or bottom flow.
• − Small spiders on the web border are slightly repetitive.

Vidu Q2

• + Strong gothic aesthetic for the thorn and branch border.
• + Good color contrast between the parchment and the jack-o-lantern.

• − Significant spelling errors in the title and subtitle text.
• − Incorrect date (30.70.2025) and time formatting.
• − Illustration style is less cinematic and feels more like a flat graphic compared to the prompt's request.

Grok Imagine Image Pro

• + Excellent preservation of the source image's facial features and clothing.
• + The hairstyle is very grounded and realistic, matching the character's age and style.
• + Seamless integration with the existing lighting and background.

• − The hair replacement slightly alters the shape of the forehead compared to the original.

Vidu Q2

• + Successfully adds a full head of hair with thick texture.
• + Preserves the majority of the background and clothing.

• − The hairstyle is overly voluminous and looks slightly unnatural/AI-generated.
• − The hair rendering looks like a flat overlay in some areas, losing the fine detail seen in the original beard.
• − The forehead and hairline transition is less realistic than the other model.

Grok Imagine Image Pro

• + Excellent 3D material rendering with realistic PBR subsurface scattering on the fish.
• + Clean, professional typography that feels integrated with the scene.
• + High geometric clarity and consistent lighting across the sushi pieces.

• − The 'diorama base' is a bit simple, appearing more like a standard wooden tray.
• − The camera angle is slightly more tilted than a true 45° isometric view.

Vidu Q2

• + Stronger adherence to the 'diorama base' prompt with a stylized raised platform.
• + Good interpretation of the miniature cartoon aesthetic.
• + Vibrant colors and creative sushi toppings.

• − The Japanese flag icon is awkwardly floating/attached to a letter.
• − The text rendering is slightly less refined with some aliasing on edges.
• − The sushi models look slightly more 'plastic' compared to the high-quality textures in Model A.

Grok Imagine Image Pro

• + Excellent adherence to the 'caricature' style with exaggerated facial features.
• + Includes many specific details like the hockey trophy, sticks, pucks, and numerous dogs.
• + Clear, legible text that complements the theme.

• − The facial likeness to the source image is somewhat lost in the extreme exaggeration.
• − The composition feels a bit crowded with many competing elements.

Vidu Q2

• + Maintains a strong facial likeness to the original subject while still being a caricature.
• + Effectively preserves the original clothing (denim shirt and black top).
• + Clean, vibrant illustration style.

• − The hands have anatomical errors, particularly the extra finger on the right hand.
• − The background hockey rink and goal are less detailed and humorous than the studio setting in Model A.

Grok Imagine Image Pro

• + Excellent depiction of god rays and sunrise atmosphere
• + Highly detailed and consistent fur texture
• + Expressive and distinct character interactions that feel playful

• − Included two kittens instead of the single one implied by the prompt
• − The positioning of the animals feels slightly static compared to the action-oriented prompt

Vidu Q2

• + Dynamic composition with a sense of movement and tumbling
• + Great variety of colorful butterflies integrated into the scene
• + Lush and vibrant wildflower meadow with beautiful bokeh

• − Inconsistent animal anatomy, especially the extra leg/tail on the golden retriever
• − Features two retriever puppies instead of the single one requested

Grok Imagine Image Pro

• + Captures the characteristic Ghibli watercolor texture and soft edge bleeding perfectly.
• + Maintains the distinct facial expressions and poses of the original meme characters while translating them to anime.
• + Features a much softer, cohesive color palette that aligns with the 'warm, nostalgic' request.

• − Slightly less clarity in the fine line work compared to traditional cel-shaded anime.
• − Background feels a bit more abstract/blurry than typical Ghibli scenic art.

Vidu Q2

• + Strong, clean line work that mimics modern anime production styles.
• + The background architectural details are more defined and visible.
• + Accurately replicates the character positions and clothing patterns from the source.

• − Lacks the specific 'hand-painted' watercolor texture associated with the Ghibli aesthetic.
• − Colors are slightly too vibrant and digital, missing the 'pastel' and 'nostalgic' mood of the prompt.
• − The woman on the right has slightly distorted facial proportions.

Grok Imagine Image Pro

• + Excellent preservation of the subject's facial features and the dog's appearance.
• + Realistic hair movement and strand detail.
• + Leaves blend naturally with the background lighting.

• − Some leaves appear a bit small and repetitive in shape.
• − The leash handle area has slight artifacts where leaves overlap.

Vidu Q2

• + Stronger sense of depth with larger, foreground leaves.
• + Successfully recreates the hair movement while maintaining image quality.
• + Preserves the overall composition of the original photo well.

• − The color of the leaves is significantly more orange/red, which slightly shifts the seasonal feel from the original.
• − Minor distortion to the subject's left eye compared to the source.

Grok Imagine Image Pro

• + Excellent typography including the grave accent in 'Caffè'
• + Clean vector emblem style with balanced composition
• + Accurate rendering of the cloche and banner exactly as requested

• − The swirl of the steam is a bit stylized compared to a realistic vapor

Vidu Q2

• + Features a nice subtle texture on the background
• + The cloche illustration has pleasant shading

• − Severe spelling errors in the main text and banner
• − Cluttered composition with redundant and mangled text
• − Included a glass-like dome rather than a traditional metal cloche

Grok Imagine Image Pro

• + Excellent typography with perfect spelling of mission steps and astronaut names.
• + Strict adherence to the requested NASA color palette and clean flat-vector style.
• + Logical vertical flow that clearly represents the six specific stages requested.

• − The lunar module icon in step 6 is slightly generic compared to real NASA hardware.

Vidu Q2

• + Features more detailed 3D-shaded icons for the lunar module.
• + Good use of the pin icon to indicate the landing site.

• − Severe spelling hallucinations including 'ALFONCH' and 'Aeutth Orot'.
• − Failure to follow the numerical sequence of the six requested steps.
• − Cluttered composition with inconsistent icon spacing and confusing labels.