FLUX.2 [dev] Black Forest Labs vs Z-Image Turbo Alibaba

FLUX.2 [dev]

• + Excellent rendering of thick, realistic glass with appropriate refractive qualities.
• + The plant is clearly visible through the glass as requested.
• + Cinematic lighting that accurately reflects the soft window light from the left.

• − The blue sphere appears to be floating slightly above the bottom surface.
• − The glass cube has rounded interior corners that make it look more like a vase or container than a geometric cube.

Z-Image Turbo

• + Sharp, precise geometric cube shapes with clean edges.
• + Realistic texture on the red book, including wear and paper detail.
• + The blue sphere is correctly seated on the bottom surface with a reflection.

• − The plant in the background is very blurry and barely visible through the glass cube itself.
• − The lighting is somewhat flat compared to the atmosphere in Model A.

FLUX.2 [dev]

• + Excellent adherence to the 'motion blur from passing cars' prompt element.
• + Very realistic skin texture and facial features for the elderly man.
• + Superior lighting and reflections on the wet pavement creating a cinematic atmosphere.
• + Highly detailed and realistic bicycle components.

• − The bicycle frame geometry becomes slightly nonsensical near the bottom bracket.
• − The man's hands are interacting with a complex mess of cables that looks a bit cluttered.

Z-Image Turbo

• + Clearer view of the 'red bicycle' as requested.
• + Good depiction of light rain falling against the background cars.

• − Failed to include motion blur for the passing cars, which remain static.
• − The skin texture and lighting look flatter and less cinematic than requested.
• − The man appears to be just holding the bike rather than repairing it.

FLUX.2 [dev]

• + Excellent execution of the braided hair with beads as requested.
• + Highly detailed texture on leather straps and metal engravings.
• + Very lifelike eyes with realistic skin texture and scars.

• − The torch in the background is slightly blurry and less defined than in Model B.

Z-Image Turbo

• + Strong atmospheric lighting with a well-defined torch and bokeh sparks.
• + Good representation of ornate plate armor and underlayers.
• + Effective use of shallow depth of field for a cinematic look.

• − The beads in the hair look more like metallic studs or sequins rather than traditional beads.
• − The skin texture and scar details are slightly softer compared to Model A.

FLUX.2 [dev]

• + Strong adherence to all requested sections including Appetizers, Pizza, and Mains
• + Excellent variety of food photography with high realism
• + Professional two-column layout that feels like a real printable menu

• − Some text rendering is garbled (e.g., PIZZAU)
• − The currency symbols are inconsistent or placed incorrectly

Z-Image Turbo

• + Very clean 3x2 image grid with vibrant, well-lit food photography
• + Cleaner font rendering for the specific section headers
• + Good use of white space and minimalist orange accents

• − Includes a major spelling error in a primary header (PIZZA MANS)
• − Fails to include the 'Mains' section correctly, instead repeating pizza and generic headers
• − The layout is less practical for a real menu with overly large gaps

FLUX.2 [dev]

• + Excellent adherence to the 'exploded' and 'suspended' burger requirement.
• + High-quality rendering of food textures, especially the lettuce and the patty.
• + Perfect text rendering including the currency symbol and fiery glow effect.

• − The 'starburst' shape for the price is a bit irregular and looks more like a fireball.

Z-Image Turbo

• + Strong professional graphic design feel with clean typography.
• + Excellent use of a sesame seed bun which adds to the photorealistic detail.
• + Vibrant colors and a very clean starburst for the price.

• − Failed the 'exploded' prompt as the burger is almost entirely assembled.
• − The double patty was not requested, though it looks appetizing.

FLUX.2 [dev]

• + Natural and realistic chalk texture on the letters
• + Accurate spelling of all requested menu items
• + Consistent cursive/handwritten style throughout

• − Slight overlapping and smudging in the second item line

Z-Image Turbo

• + Very clean and legible layout
• + Strong contrast between the chalk and the board

• − Spelling error in 'Mustroom'
• − Font looks significantly more digital/templated than handwritten
• − Missed the word 'Chip' in the third item

FLUX.2 [dev]

• + Excellent cinematic lighting and background detail with a realistic nebula.
• + High level of texture on the astronaut's suit and the horse's fur.
• + Dynamic composition with the Earth's curve and a floating moon.

• − Failed the specific logic constraint of 'horse on top' of the astronaut.
• − Anatomical issues where the horse's front legs merge together at the chest.

Z-Image Turbo

• + Clean, well-defined subject with clear outlines.
• + Good rendering of the horse's movement and tack details.
• + Accurate reflection on the astronaut's visor.

• − Failed the specific logic constraint of 'horse on top' of the astronaut.
• − Background is very plain and lacks the cinematic quality requested.
• − Lighting on the astronaut is inconsistent with the dark void environment.

FLUX.2 [dev]

• + Excellent prompt adherence with accurate character placement and expressions.
• + Superior photographic texture in the capybara's fur and the passenger's coat.
• + The lighting is more characteristic of a night scene in Manhattan with authentic bokeh.

• − The capybara's paws look slightly more like human hands than paws.
• − The 'T' on the cap is a bit simplistic.

Z-Image Turbo

• + Clean composition with a good view of both the driver and the passenger.
• + The capybara driver's paws have a more accurate animal-like texture.
• + Good rendering of the taxi driver uniform jacket.

• − The background looks more like a generic daylight or dusk street than a vibrant Manhattan night.
• − The passenger is positioned slightly too close to the front seat, making the interior feel cramped.
• − Lower overall resolution and detail compared to the other image.

FLUX.2 [dev]

• + Perfect text rendering for all requested details, including date and location.
• + High-quality gothic aesthetic with cohesive lighting and color palette.
• + Excellent composition that feels like a professional poster design.

• − The thorns and webs are slightly less integrated into the outer border compared to the illustration.

Z-Image Turbo

• + Captures the 'torn parchment' look very effectively.
• + Spooky atmosphere with a good blend of jack-o-lantern glow and background scenery.

• − Spelling error in the location text ('The Archves' instead of 'The Arches').
• − The small banner text is placed awkwardly at the very top and lacks the requested scroll graphic.
• − The layout feels cluttered with multiple separate parchment elements.

FLUX.2 [dev]

• + Excellent preservation of the source image facial features and glasses
• + True to the request for 'full, thick' hair
• + Matches the lighting and grain of the original image perfectly

• − The hair style choice (afro) may feel mismatched with the original beard texture

Z-Image Turbo

• + Natural integration of short hair that matches the beard texture

• − Fails to preserve the subject's glasses
• − Changes the subject's facial features and bone structure significantly
• − Inaccurate to the prompt for 'full, thick' hair, providing a buzz cut instead

FLUX.2 [dev]

• + Perfectly rendered text and the correct Japanese flag icon.
• + Exquisite PBR material quality with realistic wet-look tuna and grain-level rice detail.
• + Excellent adherence to the 'small raised diorama base' and isometric perspective.

• − The text color is standard black rather than a more playful cartoon style, though it matches 'bold'.

Z-Image Turbo

• + Clean, vibrant 3D cartoon style that fits the 'miniature' aesthetic.
• + Good layout and use of a solid blue background.
• + Successful rendering of the 'JAPAN' and 'SUSHI' text.

• − Significant hallucination error: included the flag of China instead of Japan.
• − The rice texture looks like large white pebbles rather than realistic rice grains.
• − The flag icon is placed to the right of the text instead of being its own element as implied by the layout request.

FLUX.2 [dev]

• + Excellent adherence to the caricature style with exaggerated facial features.
• + Includes all elements of the prompt (TV desk, hockey rink, multiple dogs in jerseys, and cameras).
• + Maintains strong facial resemblance to the subject in the source image.

Z-Image Turbo

• + Successfully preserved most of the original photo's composition and appearance.

• − Completely failed to apply the caricature style or the job elements.
• − Only added a single, blurry dog in the background with no relation to hockey or tv anchor roles.
• − Instruction to be 'exaggerated and humorous' was ignored.

FLUX.2 [dev]

• + Excellent anatomical accuracy for all animals.
• + Superior rendering of 'god rays' and atmospheric lighting.
• + Highly detailed fur texture and realistic morning dew sparkles.

• − The animals are sitting rather than 'tumbling' as requested in the prompt.
• − Included an extra rabbit.

Z-Image Turbo

• + Captures the 'tumbling' and 'playful' motion much better than Model A.
• + Bright, vibrant colors that fit the 'joyful wholesome vibe'.
• + Good adherence to the types of animals requested.

• − Noticeable anatomical issues, such as the puppy's paw merging into the rabbit.
• − The cat's facial structure and open mouth look slightly distorted and unnatural.
• − Lower overall resolution and fine detail compared to the competitor.

FLUX.2 [dev]

• + Captures the Studio Ghibli art style perfectly with soft linework and painted textures.
• + Faithfully replicates the composition and poses of the iconic meme while transforming the medium.
• + Effectively replaces the urban background with a dreamy, flowery meadow as requested.

• − The characters look significantly younger and more generic than the source image.

Z-Image Turbo

• + Maintains the facial likeness and realistic identities of the people in the original photo.

• − Completely fails the primary instruction to transform the image into a Ghibli-inspired illustration.
• − The colors are slightly muted but the image remains a photograph with no hand-painted qualities.
• − Changes the expression of the woman on the right, losing the classic 'jealous' look.

FLUX.2 [dev]

• + Excellent adherence to the hair blowing prompt with dramatic wind effect
• + High source preservation of the background and the dog
• + Effective addition of numerous flying leaves.

• − The hair edges look slightly artificial against the background.

Z-Image Turbo

• + Natural appearance of the subject's face
• + Subtle wind effect on the hair is more realistic.

• − Failed to significantly change the hair appearance compared to the source
• − Fewer flying leaves than Model A
• − Significant changes to the background elements like the flowers and river bank.

FLUX.2 [dev]

• + Excellent typography integrated into a complex banner design.
• + Strong vector emblem aesthetic with detailed hatching and subtle textures.
• + Faithfully includes all prompt elements including the cloche and 'Est. 1720' banner.

• − The 'E' in 'Caffè' has a slightly irregular accent mark style.

Z-Image Turbo

• + Clean, minimalist presentation with high legibility.
• + Accurate text rendering and placement.
• + Soft, pleasing background texture.

• − Failed to include the specific 'banner' requested for the 'Est. 1720' text.
• − Simplified vector art lacks the 'vintage' character seen in the rival image.

FLUX.2 [dev]

• + Excellent adherence to the chronological step-by-step layout requested.
• + High-quality vector aesthetic with consistent lighting and iconography.
• + Correctly identifies names like Armstrong, Aldrin, and Collins in the header.

• − Includes several garbled text labels and redundant icons that weren't in the prompt.
• − The layout is a bit cluttered with nine icons instead of the six requested steps.

Z-Image Turbo

• + Clean white background following the 'modern vector' style effectively.
• + The 'Earth Orbit' and 'Moon' icons are simple and readable.
• + Captures the specified NASA-inspired color palette well.

• − Major spelling errors in the title ('APOLIO E 11') and labels ('Descenty', 'Translurian').
• − Fails to show the sequence of steps, missing several instructed icons like the trajectory arc.
• − Composition feels empty and lacks the professional finish of an infographic.