FLUX.1 Kontext [max] Black Forest Labs vs GPT Image 1.5 OpenAI

FLUX.1 Kontext [max]

• + Excellent handling of light and shadows, creating a realistic depth.
• + Accurately places the plant behind the cube as requested.
• + Realistic textures on the wooden table and sparking sphere.

• − The sphere appears to have a textured or glittery surface rather than a smooth one.
• − The glass cube looks more like a shallow container with an open top where the book rests.

GPT Image 1.5

• + Clearer representation of a 'glass cube' structure.
• + Good adherence to all spatial instructions in the prompt.
• + The blue sphere is smooth and geometrically perfect.

• − The plant appears to be inside the cube's reflections rather than logically placed behind it.
• − Lighting is a bit flat compared to the other model.

FLUX.1 Kontext [max]

• + Excellent depiction of rain streaks that feel integrated into the 3D space.
• + Strong cinematic lighting with realistic reflections on the pavement.
• + Captures a very natural, aged skin texture on the subject's hands and face.

• − The anatomy of the bicycle is highly distorted, specifically the chain and crankset area.
• − The subject does not look distinctly Japanese, appearing more generically elderly.

GPT Image 1.5

• + Much better bicycle anatomy including a realistic chain, gear set, and accessories like a basket.
• + The subject perfectly matches the demographic requested in the prompt.
• + The 'imperfect framing' and composition feel more like a genuine candid street photo.

• − The rain effect is very subtle, almost difficult to see compared to the puddles.
• − Slightly less 'cinematic' lighting compared to the first model.

FLUX.1 Kontext [max]

• + Excellent metal texture with realistic engraving and dirt patterns.
• + Strong dramatic lighting that creates a moody atmosphere.

• − The hair braids are thick and lack the requested beads.
• − Skin texture looks slightly overly smooth/metallic compared to the armor.

GPT Image 1.5

• + Perfect adherence to the 'beads in braids' and 'faint scars' prompt elements.
• + Highly detailed texture on the leather straps and multiple layers of clothing.
• + Excellent facial detail and lifelike expression with realistic dirt smudges.

• − The depth of field is slightly less pronounced than Model A's composition.
• − The warm torchlight highlights are a bit harsh in some areas of the plate.

FLUX.1 Kontext [max]

• + Strong minimalist aesthetic with a clean white professional layout
• + Food photography is well-integrated and looks high-quality
• + Excellent use of negative space for a modern brand feel

• − Text is mostly gibberish or contains severe misspellings
• − Does not follow the grid structure for food photos as strictly
• − Contrast is a bit low on some font elements

GPT Image 1.5

• + Perfect text rendering with realistic restaurant item names and prices
• + Matches the specific grid layout requested in the prompt
• + Clearly defines sections for Appetizers, Pizza, and Mains as requested

• − Layout is slightly more generic and less stylish than Image A
• − The horizontal lines between items feel a bit heavy for a 'minimalist' design

FLUX.1 Kontext [max]

• + Excellent text legibility and clean graphic design.
• + High photorealistic detail on the burger patty and lettuce texture.
• + Sophisticated lighting on the environment and food.

• − The main burger is mostly assembled rather than properly 'exploded'.
• − Missing the starburst element for the price.
• − Strange floating orange bun-like shapes on the sides.

GPT Image 1.5

• + Perfectly follows the 'exploded' mechanical instruction with distinct layers.
• + Includes all requested text elements, including the starburst and fiery effect.
• + Strong sense of motion and energy with food and sparks.

• − The 'MAGIC BURGER' text is slightly cut off at the top.
• − Over-sharpened visual style that looks more like a composite than a single photo.

FLUX.1 Kontext [max]

• + Excellent text legibility and spelling accuracy
• + Realistic chalkboard smudges and background details of the café Environment
• + Includes additional appropriate context-related text at the bottom

• − The handwriting style is a bit too clean and uniform, bordering on look-alike font
• − Missing the requested 'elegant cursive' for the title line

GPT Image 1.5

• + Successfully captured the 'elegant cursive' style for the title as requested
• + Highly realistic chalk texture with authentic varying opacity and dustiness
• + Handwriting variations feel more organic and less like a digital font

• − Text is slightly less legible due to the heavy chalk grain
• − The background is just a flat board with less environmental context than Model A

FLUX.1 Kontext [max]

• + Excellent anatomical rendering of the horse and astronaut.
• + Clean, cinematic lighting with a minimalist and surreal background.
• + High-fidelity textures on the spacesuit and horse's coat.

• − Fails to follow the tricky 'horse on top' spatial instruction.
• − Background is slightly generic compared to the detailed foreground.

GPT Image 1.5

• + Rich, busy composition with multiple celestial elements like Saturn and a lunar lander.
• + Dynamic sense of motion with the dust kicking up on the lunar surface.
• + Strong cinematic atmosphere with complex lighting.

• − Fails to follow the 'horse on top' spatial instruction.
• − Anatomical issues with the horse's front legs and hooves.

FLUX.1 Kontext [max]

• + Excellent photorealistic texture and lighting
• + Clean, high-resolution details on the capybara's fur and the taxi cap
• + The composition creates a cinematic depth of field

• − The passenger is holding a phone to her ear like a call instead of looking at it as requested
• − Only one paw is clearly on the steering wheel

GPT Image 1.5

• + Perfect adherence to the pose prompt with both paws on the steering wheel
• + Captures the bored expression of the passenger looking at her phone accurately
• + Authentic taxi interior details like the meter and dashboard

• − Image quality is slightly grainier compared to the alternative
• − Clothing textures on the capybara are a bit muddy

FLUX.1 Kontext [max]

• + Excellent font choice that feels authentically gothic
• + Highly legible text with strong contrast
• + Clever integration of cobwebs into the corners

• − Redundant text at the bottom repeats the location and date differently
• − The 'parchment' texture is very subtle and looks more like a modern poster
• − Jack-o-lantern rendering is a bit flat compared to the background

GPT Image 1.5

• + Beautiful sepia-toned parchment aesthetic with rich textures
• + Excellent composition with a detailed graveyard background and moon
• + More intricate border work featuring thorns and webs

• − Minor text artifacts in the cursive banner
• − The main title font is slightly less 'gothic' than Model A

FLUX.1 Kontext [max]

• + Satisfies the prompt for a full, thick head of hair
• + Maintains original lighting and jacket details
• + Natural hair color matching the beard

• − Noticeably alters the man's facial features and eye shape
• − Hairline and sideburn integration looks slightly unnatural or layered on top
• − Changed the frame color and thickness of the glasses

GPT Image 1.5

• + Excellent preservation of original facial features and expression
• + Highly realistic hair texture and natural integration with the existing sideburns
• + Perfectly maintains the background and accessories like the glasses

• − The hair density is slightly less 'full' compared to Model A, though still realistic

FLUX.1 Kontext [max]

• + Excellent soft 3D cartoon style consistent with the 'miniature' prompt.
• + Clean, minimalist composition that adheres to the specific texture requests.
• + Accurate text rendering for both requested words.

• − Missed the request for a small flag icon.
• − The isometric angle is slightly shallow compared to a standard 45° view.

GPT Image 1.5

• + Successfully included all elements, including the flag icon.
• + Outstanding PBR material rendering, especially on the wood and ceramic surfaces.
• + Perfect 45° isometric perspective.

• − More cluttered than the requested 'minimal garnish' scene.
• − The text style is a bit flatter compared to the 3D aesthetic of the objects.

FLUX.1 Kontext [max]

• + Maintains the subject's denim clothing from the source image
• + Clear caricature art style with bold lines
• + Good integration of the hockey stick in the background

• − Added glasses which were not in the source image
• − The caricature features look generic and less like the specific person
• − The dog is small and somewhat obscured in the background

GPT Image 1.5

• + Exaggerated features strongly resemble the subject's face
• + Highly detailed news studio environment with a professional broadcast look
• + Creative and humorous integration of hockey (the screen and the small dog in a helmet)

• − Swapped the denim shirt for a red dress
• − The large dog's left paw/leg is anatomically messy
• − The text on the lower third is slightly stylized but readable

FLUX.1 Kontext [max]

• + Excellent adherence to the 'golden sunrise' light with beautiful god rays
• + Extremely clean and high-definition fur textures on all animals
• + Well-composed and organized group portrait that includes all requested animals clearly

• − The animals are largely static and sitting rather than 'tumbling and chasing' as requested
• − The lighting on the butterfly wings feels a bit flat compared to the backgrounds

GPT Image 1.5

• + Perfectly captures the 'tumbling' and 'playfully chasing' action described in the prompt
• + Captures the 'dew sparkles' effect very effectively across the meadow
• + Dynamic poses create a more wholesome and joyful vibe

• − The fox kit has structural issues with its anatomy, specifically the placement of its paws and neck
• − The kitten's pose is a bit chaotic and slightly distorts its facial features

FLUX.1 Kontext [max]

• + Excellent preservation of the original composition and poses.
• + Captures the iconic hand-painted watercolor texture of Studio Ghibli backgrounds.
• + Maintains clear facial expressions that match the source image context.

• − The character design looks more like modern TV anime rather than the specific Soft Ghibli style.

GPT Image 1.5

• + Successfully captures the 'warm, nostalgic mood' with a beautiful golden-hour lighting effect.
• + Effective use of soft focus and pastel tones to create a dreamy atmosphere.
• + Preserves the original arrangement of characters perfectly.

• − The woman in the foreground is excessively blurred, losing too much detail.
• − The lighting effect, while pretty, borders on a generic 'shoujo' filters rather than the clean look of Ghibli.

FLUX.1 Kontext [max]

• + Successfully added wind-blown hair effect
• + Includes leaves flying in the air as requested
• + Adds a dynamic gait to the walking pose

• − Anatomy of the left hand is significantly distorted
• − Changes the background and overall composition more than necessary

GPT Image 1.5

• + Excellent depiction of wind-blown hair that looks natural
• + Abundant flying leaves create a high sense of energy
• + Preserves the original appearance of the person and dog very accurately

• − Some leaves appear to be floating or lack motion blur, reducing the realism of the wind

FLUX.1 Kontext [max]

• + Perfectly captures the 'light background' and 'subtle texture' requested in the prompt.
• + Excellent typography with accurate spelling of 'Caffè' including the accent mark.
• + Clean, minimalist vector design that functions well as a logo.

• − The 'Est. 1720' text is slightly off-center within the banner.

GPT Image 1.5

• + More detailed illustration of the cloche with effective use of highlights and shading.
• + Good use of classic, elegant typography for the restaurant name.
• + Accurate banner design with correctly rendered 'Est. 1720' text.

• − Failed to follow the 'light background' instruction, providing a black background instead.
• − The typography style for 'Caffè' is slightly disconnected from the 'Florian' font choice.
• − The overall design feels more like a detailed illustration than a minimalist vector logo.

FLUX.1 Kontext [max]

• + Excellent font usage for 'APOLLO 11' title.
• + Features high-quality silhouettes of the crew members.
• + Strong flat-vector aesthetics that feel cohesive.

• − Confusing informational flow with arrows pointing in illogical directions.
• − Labels do not match the icons shown (e.g., 'Translunar' is under a moon icon).
• − Contains significant gibberish text in the center.

GPT Image 1.5

• + Perfect adherence to the 6-step logical sequence requested.
• + Very clean, legible text and clearly labeled sections.
• + Consistency in iconography across all six stages of the mission.

• − Spacecraft designs are slightly more generic and less detailed than Image A.
• − The 'Saturn V' rocket in the launch phase is a bit simplified.