FLUX.1 Kontext [max] Black Forest Labs vs Vidu Q2 ShengShu Technology

FLUX.1 Kontext [max]

• + Excellent handling of caustics and light refraction through the glass.
• + Sophisticated photographic realism with soft, natural depth of field.
• + Accurate placement of all requested elements.

• − The blue sphere has a textured, glittery appearance rather than a smooth finish.
• − The plant in the background is quite dark and slightly obscured.

Vidu Q2

• + Strong prompt adherence with vibrant, clear colors.
• + The glass cube has realistic thickness and greenish edge tinting typical of glass.
• + Excellent visibility of the green plant through the glass panels.

• − The book appears slightly small relative to the cube.
• − The lighting is a bit harsh, creating high-contrast shadows that feel less 'soft'.

FLUX.1 Kontext [max]

• + Excellent depiction of rain with realistic streaks and a hazy atmosphere.
• + Strong cinematic composition with a centered subject and convincing bokeh.
• + Naturally rendered skin textures and hair on the subject.

• − The bike chain and mechanics are physically impossible, merging into the wheel.
• − The rain streaks appear layered over the lens rather than interacting with the 3D space.

Vidu Q2

• + Effectively captures a candid feel with 'imperfect' close-up framing.
• + Better representation of motion blur from the passing car in the background.
• + Detailed skin textures and realistic hand anatomy.

• − The main bicycle structure is nonsensical, having two front forks and two front wheels.
• − Does not show 'light rain' falling in the air, only wet surfaces.
• − The bicycle chain is disconnected and floating near the pedal.

FLUX.1 Kontext [max]

• + Excellent close-up skin texture and realistic depth of field
• + Strong adherence to the 'battle-worn' descriptor through rugged features
• + Very high detail in the metallic engravings

• − The eyes appear slightly glowing/supernatural rather than just 'lifelike'
• − Missing the beads in the braids requested by the prompt

Vidu Q2

• + Perfect inclusion of beads in the braided hair
• + Excellent portrayal of scars, dirt, and blood on the skin
• + Complex and highly detailed leather straps and cloth underlayers

• − The composition is a medium shot rather than the 'close portrait' requested
• − The skin texture on the face looks slightly smoother and more digital compared to Model A

FLUX.1 Kontext [max]

• + Excellent structure that favors a realistic menu layout
• + Clean and professional header with a coherent logo
• + High-quality, appetizing food photography

• − Repetitive food images (mostly pizza) despite prompt for variety
• − Smaller body text is largely illegible gibberish
• − The script font used for section headers is difficult to read

Vidu Q2

• + Includes all requested categories like Mains and Appetizers clearly labeled
• + Uses bold sans-serif fonts as requested
• + Vibrant food photography with good color balance

• − Layout feels somewhat cluttered and chaotic
• − Text contains many typos and nonsensical word fragments
• − Borders and graphic accents are a bit inconsistent

FLUX.1 Kontext [max]

• + Excellent typography rendering and spacing
• + Highly realistic textures on the patty and bun

• − The burger is not truly 'exploded' as it remains largely assembled in the center
• − Missing the requested starburst element for the price

Vidu Q2

• + Successfully creates the 'exploded' look with vertically separated layers and flying sauce
• + Strong adherence to all prompt elements including the starburst and fiery background
• + Dynamic energy and motion feel more pronounced

• − Minor text artifact in the price where the currency symbol is distorted
• − The lighting on the lettuce looks a bit flat compared to the fiery surroundings

FLUX.1 Kontext [max]

• + Perfect text accuracy for all requested menu items and dates
• + Authentic board texture with realistic chalk smudges and dust
• + Clean, legible handwriting that maintains a natural chalk appearance

• − The title is in print-style block letters rather than the requested elegant cursive

Vidu Q2

• + The title uses a more cursive/fluid style as requested in the prompt
• + Excellent chalk texture with realistic color variations and pressure depth

• − Severe spelling errors throughout the menu items such as 'Truffe Musshoom' and 'Lemepun'
• − The text becomes unintelligible toward the bottom of the board

FLUX.1 Kontext [max]

• + Excellent anatomical rendering of the horse and astronaut gear.
• + Cinematic lighting with a realistic light source from a distant planet.
• + High-fidelity textures on the horse's fur and the suit material.

• − The composition is a bit static and conventional.

Vidu Q2

• + Strong surrealist interpretation with the nebula-skin horse.
• + Dynamic and colorful composition that feels more imaginative.
• + Excellent use of the background galaxy to frame the subject.

• − The astronaut's face is visible through the visor but lacks detail.
• − The horse's anatomy is slightly distorted in the rear legs.

FLUX.1 Kontext [max]

• + Excellent photorealistic fur texture and lighting on the capybara.
• + Successfully includes 'TAXI' text on the driver's cap.
• + High-quality blurred background bokeh that feels like a real city night.

• − The passenger is holding a phone to her ear like a call, rather than looking at it as requested.
• − The perspective chosen makes it difficult to see both paws clearly on the steering wheel.

Vidu Q2

• + Perfectly captures the passenger looking at her phone with a bored expression.
• + Clearly shows both paws on the steering wheel as requested.
• + Strong composition that shows more of the taxi interior and the street outside.

• − The capybara's hands look slightly more humanoid/primate-like than actual capybara paws.
• − The lighting on the capybara's face is a bit flat compared to Model A.

FLUX.1 Kontext [max]

• + Perfect text accuracy for all requested fields
• + Excellent atmospheric lighting and cohesive color palette
• + Professional and clean composition formatted perfectly for an invitation

• − The date format uses commas instead of dots
• − Repeated the location text in the bottom section

Vidu Q2

• + Stronger parchment texture and torn edge effect
• + Higher contrast in the background elements like thorns and webbing
• + Captures the vintage aesthetic effectively

• − Significant spelling errors in the title, banner, and date
• − The location and time details are merged and include gibberish text
• − Lighting on the pumpkin feels less integrated with the environment

FLUX.1 Kontext [max]

• + Successfully added a thick, full head of hair that matches the prompt.
• + Maintains high resolution and sharp details throughout the image.

• − Significantly altered the man's facial features, making him look like a different person.
• − The hairline looks slightly artificial and lacking in fine transitional hairs.

Vidu Q2

• + Excellent preservation of the original man's identity and facial features.
• + Realistic hair texture and a very convincing, natural-looking hairline.
• + Perfectly preserved the original lighting, clothing, and background.

• − Small stray hairs on the right side of the head appear slightly digital/stretched when viewed closely.

FLUX.1 Kontext [max]

• + Excellent typography with a clean, playful 3D font style.
• + Superior material rendering for the nigiri and plate.
• + Very clean composition that perfectly follows the diorama base request.

• − Failed to include the requested flag icon.
• − The chopsticks are merged into the board surface slightly awkwardly.

Vidu Q2

• + Includes all requested elements including the flag icon.
• + Clean and legible text rendering.
• + Good variety of sushi types on the plate.

• − Lighting is a bit harsh and flat compared to Model A.
• − The textures look more like generic plastic and lack the 'soft refined' quality.
• − Visual artifacts present on the edges of the diorama base.

FLUX.1 Kontext [max]

• + Successfully incorporates all elements: TV anchor desk, a dog, and a hockey stick.
• + Maintains the denim shirt and black top from the source image effectively.
• + Strong caricature style with exaggerated facial features and clean line art.

• − The hockey stick is barely visible and cut off at the top edge.
• − The person's face looks significantly different from the source image, losing some identity recognition.
• − Randomly added glasses that weren't in the source image.

Vidu Q2

• + Excellent preservation of the subject's facial features while still achieving a caricature effect.
• + Superb integration of the hockey theme with a rink background, goal, and puck.
• + High visual variety including two different dog breeds and paper with paw prints.

• − The hand holding the microphone has anatomical issues with finger placement.
• − The left hand (resting on a stack of papers) has an extra-long ring finger and simplified anatomy.

FLUX.1 Kontext [max]

• + Excellent fur texture rendering and soft lighting.
• + Captures the 'big expressive eyes' and 'wholesome vibe' perfectly.
• + Includes all four distinct animals requested.

• − The animals are sitting still rather than 'playfully chasing and tumbling'.
• − The butterflies appear somewhat flat and lack detail.

Vidu Q2

• + Excellent adherence to the 'playfully chasing' and 'tumbling' part of the prompt.
• + Superior butterfly detail and variety.
• + Dynamic composition with a strong sense of movement.

• − Duplicated the golden retriever puppy, including two instead of one.
• − The rabbit has a somewhat unusual spotted pattern and tail that looks slightly hybrid.
• − Artifacts on the puppy's paws in the background.

FLUX.1 Kontext [max]

• + Excellent adherence to the Ghibli art style with soft, watercolor-like textures
• + Perfectly captures the 'warm, nostalgic mood' requested in the prompt
• + Preserves the composition and poses of the original meme perfectly

• − The faces look a bit generic and lose some of the specific expressions from the source

Vidu Q2

• + Successfully translates the image into an illustrative style
• + Retains more of the original character features, especially in the man's face
• + Good use of pastel colors and soft lighting

• − The line work is a bit too clean and lacks the hand-painted, textured feel of Ghibli backgrounds
• − The background characters are slightly more distorted than in Image A

FLUX.1 Kontext [max]

• + Excellent wind physics applied to the hair and the dog's tail
• + Subtle but effective placement of small green leaves matching the local trees
• + Changes to the pose, like the arm outreach, enhance the energetic feel

• − Anatomical error with the left hand being slightly mangled
• − Significant changes to the appearance of the face compared to the source image

Vidu Q2

• + Strong adherence to the 'flying leaves' request with vibrant autumn colors
• + Excellent preservation of the woman's facial features and the overall scene structure
• + Good motion effect in the hair

• − The orange leaves feel a bit disconnected from the summer-green trees in the background
• − Less dynamic change to the character's posture compared to Model A

FLUX.1 Kontext [max]

• + Excellent typography with perfect spelling of 'Caffè Florian'.
• + Authentic vintage woodblock/vector style with appropriate texture.
• + Clean, balanced composition that follows all prompt instructions.

• − The steam is very minimalist, bordering on abstract.

Vidu Q2

• + Elegant warm cream and gold color palette.
• + Good illustration of the cloche dome.

• − Severe spelling errors including 'Caffe Farmiin' and 'Fopli20'.
• − Redundant and cluttered text layout.
• − Illogical steam placement inside and outside the glass.

FLUX.1 Kontext [max]

• + Excellent vector illustration style that feels professional and cohesive.
• + Matches the requested navy, muted red, and light gray palette perfectly.
• + Stronger execution of complex graphical elements like trajectory arcs and the lander.

• − Internal labels are confused (e.g., Earth is labeled 'Moon').
• − Includes Saturn as a background element which was not requested.
• − The 'Saturn V' looks more like a 1950s atomic rocket than the actual NASA vehicle.

Vidu Q2

• + Follows the step-by-step layout much more logically than Model A.
• + Includes better iconography for Earth and Lunar orbits.
• + Maintains a cleaner, more minimalist infographic aesthetic.

• − Text rendering is mostly gibberish (e.g., 'ALFONCH' instead of APOLLO).
• − The 'Saturn V' icon is very generic and lacks detail.
• − Visual quality is slightly softer/lower resolution compared to Model A.