FLUX.1 Kontext [max] Black Forest Labs vs Z-Image Turbo Alibaba

FLUX.1 Kontext [max]

• + Excellent handling of light and reflections, including caustic patterns on the table.
• + High level of detail in the textures of the wood, book, and sphere.
• + Very accurate spatial reasoning with the plant correctly refracted through the glass.

• − The text on the book spine is nonsensical gibberish.
• − The blue sphere has an unusually rough, glittery texture instead of a smooth finish.

Z-Image Turbo

• + Clean, simple composition that matches all primary prompt elements.
• + The blue sphere has a realistic, smooth glass-like quality.
• + Accurate lighting from the left following the prompt's direction.

• − Lower overall resolution and slightly blurry textures on the plant and background.
• − The plant is simple and lacks the detail and realistic leaf structure shown in Model A.
• − Refractions through the glass are less complex and photorealistic.

FLUX.1 Kontext [max]

• + Excellent adherence to the 'candid' and 'imperfect framing' prompt with a low-angle perspective.
• + Superior rendering of rain droplets and light reflections on the pavement.
• + Captures the cinematic atmosphere and shallow depth of field perfectly.

• − The bike chain and spoke details are physically impossible (entwined incorrectly).

Z-Image Turbo

• + Good rendition of natural skin texture on the man's arms and face.
• + The subject is clearly Japanese and fits the 'elderly' description well.

• − Failed to include 'motion blur from passing cars' as the background cars are sharp.
• − Missed the 'repairing' action, as the man appears to be simply standing with or holding the bike.
• − Lacks the cinematic lighting and wet pavement reflections requested.

FLUX.1 Kontext [max]

• + Exceptional macro detail on the engraved plate armor and skin texture
• + Strong adherence to the 'close portrait' instruction with a powerful composition
• + High-quality lighting consistency with golden torchlight reflecting across the face and metal

• − Missed the 'hair braided with small beads' detail, showing hanya thick braids
• − Skin texture appears slightly over-sharpened and excessively sweaty

Z-Image Turbo

• + Includes the 'small beads' in the hair braids as specifically requested
• + Good variety of textures including chainmail, leather, and engraved plate
• + Atmospheric composition with a visible torch source and bokeh

• − The 'close portrait' is more of a medium shot, losing some facial detail
• − The fire source appears slightly detached from the hand/torch handle

FLUX.1 Kontext [max]

• + Clean, professional layout with a focus on whitespace
• + Accurate representation of a full menu sheet including pricing and logo
• + High-quality, appetizing food photography

• − Text is mostly illegible gibberish
• − The grid layout isn't as distinct for the food photos

Z-Image Turbo

• + Strong adherence to the 'grid' request for photos
• + Better text rendering with clear headers like 'Appetizers' and 'Pizza'
• + Vibrant orange accents create a cohesive visual style

• − Layout is a bit cluttered and lacks formal menu hierarchy
• − Misspelling of 'Mains' as 'Mans'
• − Image proportion is square rather than a standard vertical menu

FLUX.1 Kontext [max]

• + Excellent photorealistic texture on the meat and buns
• + Strong sense of motion with exploding ingredients
• + Highly legible and accurately integrated text

• − The burger components are partially assembled rather than fully exploded horizontally or vertically
• − The starburst for the price is very small and lacks impact

Z-Image Turbo

• + Perfect adherence to the starburst requirement for the price
• + Vibrant glowing effect on the text and assets
• + Clean and professional graphic design layout

• − Failed the 'exploded burger' requirement, showing a mostly assembled burger
• − Less photorealistic texture compared to the competitor
• − Background feels more like a studio set than a fiery void

FLUX.1 Kontext [max]

• + Excellent text accuracy with no spelling errors
• + Includes realistic chalk smudges and texture on the board
• + Successfully completed the truncated prompt text with 'Cookies'

• − The heading is in sans-serif print rather than cursive as requested
• − Text style is a bit too uniform, bordering on marker style rather than dusty chalk

Z-Image Turbo

• + Beautiful chalk texture with realistic variation in thickness and opacity
• + Captures a hand-drawn feeling with better character spacing and slant

• − Includes a spelling error with 'Mustroom' instead of 'Mushroom'
• − Failed to include 'elegant cursive' for the title as requested

FLUX.1 Kontext [max]

• + Excellent photographic lighting and cinematic atmosphere
• + High detail in the horse's fur and the astronaut's suit textures
• + Beautiful composition with the curve of the planet in the background

• − Failed the specific spatial instruction for the horse to be on top of the astronaut

Z-Image Turbo

• + Clean, sharp image quality
• + Good anatomical rendering of the horse
• + Consistent lighting on the subjects

• − Failed the specific spatial instruction for the horse to be on top of the astronaut
• − Background is relatively flat and less cinematic compared to Model A

FLUX.1 Kontext [max]

• + Excellent fur texture rendering and realistic lighting on the capybara.
• + Captures a very cinematic, high-quality photorealistic aesthetic.
• + The passenger is correctly placed in the background with a phone.

• − Only one paw is clearly visible on the steering wheel instead of both requested.
• − The passenger appears to be making a call rather than just looking at the screen.

Z-Image Turbo

• + Better adherence to the 'both front paws on the steering wheel' instruction.
• + The passenger's expression and pose perfectly match the 'bored businesswoman looking at phone' prompt.
• + Realistic integration of the animal's limbs with the steering wheel.

• − The lighting is a bit flat compared to the cinematic quality of the competitor.
• − The capybara's fur looks slightly less detailed and soft than in the other image.

FLUX.1 Kontext [max]

• + Excellent typography with a dark, polished gothic aesthetic.
• + Perfect adherence to the dark parchment and cinematic lighting request.
• + Accurate text rendering for all lines including the secondary banner.

• − Repetitive location text at the bottom.
• − The layout is a bit cluttered with the text overlapping the dark background elements.

Z-Image Turbo

• + Strong composition with a clear, readable scroll-based design.
• + Good inclusion of all requested elements like thorns, webs, and twisted trees.
• + Vibrant glow on the jack-o-lantern.

• − Spelling error in the location ('The Archves' instead of 'The Arches').
• − The parchment looks a bit too bright and clean for a 'dark gothic' theme.
• − Secondary text at the top is very small and plain.

FLUX.1 Kontext [max]

• + Excellent adherence to the 'full, thick head of hair' request.
• + Maintains very high visual consistency with the original background and lighting.
• + The texture and volume of the hair look natural and integrated.

• − The facial features were slightly altered, making the person look like a younger/different version of the source.

Z-Image Turbo

• + The facial identity is preserved slightly better in terms of bone structure.

• − Failed the primary instruction to add a full head of hair, providing only a slight buzz-cut shadow.
• − Significant change to the background scenery (added grass and mountains).
• − Removed the subject's glasses entirely.

FLUX.1 Kontext [max]

• + Excellent typography style that matches the 3D aesthetic.
• + High-quality 3D renders with realistic textures on the fish and rice.
• + Captures the diorama feel well with the addition of chopsticks and wasabi.

• − Missing the requested flag icon.
• − The perspective is slightly lower than a true 45-degree isometric view.

Z-Image Turbo

• + Successfully included a flag icon as requested.
• + Clean, minimalist composition that adheres strictly to the 'small diorama base' request.
• + Accurate isometric perspective.

• − Used the flag of China instead of Japan, which is a major hallucination/error given the text 'JAPAN'.
• − The single piece of sushi is less visually interesting than the pair in the other model.

FLUX.1 Kontext [max]

• + Successfully creative and fun caricature style consistently applied.
• + Includes all requested elements like the hockey stick, dog, and news desk.
• + Maintains character likeness through hair and outfit color.

• − The hockey stick is positioned awkwardly in the top corner.
• − Complete departure from the original photograph's realistic style.

Z-Image Turbo

• + Preserves the source face and photographic style with high fidelity.
• + Subtly adds a small dog in the background area.

• − Fails to apply the core caricature request.
• − Misses the television anchor and hockey themes entirely.
• − The added dog is very small and lacks detail.

FLUX.1 Kontext [max]

• + Excellent depiction of warm golden light and volumetric god rays
• + High consistency in the 'lush meadow' environment with vibrant wildflowers
• + Strong adherence to the 'butterflies' prompt with multiple elements integrated into the scenery

• − The animals are sitting relatively still rather than 'playfully chasing' as requested
• − The bunny has a slightly artificial, cartoonish expression compared to the others

Z-Image Turbo

• + Better captures the 'playfully chasing' and 'tumbling together' action requested in the prompt
• + Superior fur texture and realism on the golden retriever and fox
• + Naturalistic dew sparkles on the grass in the foreground

• − Lighting is less dramatic, lacking the requested 'god rays' effect seen in the other model
• − Butterflies appear a bit static and disconnected from the animals' eyelines

FLUX.1 Kontext [max]

• + Perfectly captures the Studio Ghibli art style with hand-painted watercolor textures.
• + Excellent preservation of the source image's composition and the 'distracted boyfriend' meme logic.
• + Successfully translates the character expressions and poses into a nostalgic anime aesthetic.

• − The woman in the red dress is now in sharp focus, which differs from the bokeh effect in the original image.

Z-Image Turbo

• + Maintains the realistic depth of field from the source image.

• − Completely failed to apply the Ghibli illustration style, resulting in a slightly filtered photo rather than art.
• − The character faces are altered to new people rather than artistic interpretations of the originals.
• − Lacks the requested soft pastel colors and hand-painted textures.

FLUX.1 Kontext [max]

• + Excellent adherence to the 'windy hair' prompt with natural-looking flow.
• + Successfully added flying leaves throughout the scene.
• + Strong preservation of the original subjects' faces and clothing style.

• − Minor anatomic issues with the woman's left hand hovering unnaturally near the dog's head.
• − The leash has become detached and floats in a loop.

Z-Image Turbo

• + Effectively added falling and flying leaves to the composition.
• + Maintains a high level of visual quality and consistency with the source lighting.

• − The hair shows very little motion compared to Model A, failing that part of the prompt.
• − Noticeable change to the woman's facial features, losing the likeness of the source image.
• − The woman's left hand is poorly rendered and appears to be missing most fingers.

FLUX.1 Kontext [max]

• + Excellent typography with proper accented characters and vintage texture
• + Accurately follows the banner request for the 'Est. 1720' text
• + Provides a genuine hand-drawn vintage feel with consistent woodblock-style texture

• − Slightly less 'minimalist' than a modern corporate logo, though fitting for the vintage brief

Z-Image Turbo

• + Clean vector-style execution
• + Good color palette adherence
• + Legible typography

• − Missed the 'banner' element for the date, using horizontal lines instead
• − The cloche icon is very generic and lacks the requested vintage detail
• − Overall appearance looks like a modern digital imitation rather than an authentic vintage emblem

FLUX.1 Kontext [max]

• + Excellent typography and spelling throughout the graphic.
• + Sophisticated composition with a clean, professional NASA-inspired color palette.
• + Included creative additions like the correctly identified crew member silhouettes.

• − Failed to include exactly 6 steps as requested in the sequence.
• − The rocket icon looks more like a generic cartoon rocket than a Saturn V.
• − Logic flow of the arrows and labels is slightly confusing.

Z-Image Turbo

• + Successfully captured the requested flat-vector style with crisp icons.
• + Followed the color palette requirements accurately.
• + The rocket design more closely resembles the staging of a heavy-lift launch vehicle.

• − Multiple spelling errors including 'APOLIO E 11', 'Translurian', and 'Descenty'.
• − Poorly aligned text and floating elements create a messy composition.
• − Missing some of the requested steps in the infographic sequence.