GPT Image 1.5 OpenAI vs LongCat-Image Meituan

GPT Image 1.5

• + Excellent photographic texture on the red book cover and pages.
• + Superior representation of plants visible through the glass with realistic distortion.
• + Perfect adherence to all spatial requirements in the prompt.

• − The sphere is quite large relative to the 'small' descriptor.
• − The bottom edge of the glass cube looks slightly thick, almost like a tray.

LongCat-Image

• + Beautiful translucent material for the blue sphere.
• + Clear lighting direction from the window as requested.
• + Clean, minimalist composition.

• − The glass refractive logic is a bit weak, as the plant behind it looks almost unchanged through the glass.
• − The book's placement and texture feel slightly more synthetic than Model A.

GPT Image 1.5

• + Excellent handling of wet textures and reflections on the pavement.
• + Captures highly realistic skin texture and clothing details with a candid feel.
• + Accurate bike mechanics and authentic rain particles.

• − The car in the background lacks the specific motion blur requested in the prompt.

LongCat-Image

• + Successfully incorporates motion blur into the passing cars.
• + Good use of cinematic lighting from headlights and street storefronts.
• + Matches the 'imperfect framing' request by including a foreground pole.

• − The red bicycle involves nonsensical geometry with three wheels and overlapping frames.
• − The man's hands and the bike parts have significant AI artifacts.
• − The rain looks like vertical streaks rather than a natural mist.

GPT Image 1.5

• + Exquisite micro-textures on the engraved metal and worn leather straps.
• + Masterful use of lighting and depth of field that feels professional and cinematic.
• + Authentic portrayal of 'battle-worn' with realistic dirt and grimy skin texture.

• − The hair beads are somewhat subtle and blend into the hair color.

LongCat-Image

• + Clearly follows the request for braided hair with visible beads.
• + Good contrast between the polished armor and the fire effects in the background.
• + Correctly incorporates the cloth underlayer and leather straps.

• − The scarring looks like a digital stamp rather than a natural part of the skin.
• − The lighting on the character feels flat and disconnected from the bright fire behind them.
• − Overall image has a slightly smoothed, CGI aesthetic compared to the first.

GPT Image 1.5

• + Excellent typography with perfect spelling and pricing consistency.
• + Clean, professional layout that matches the 'modern minimalist' aesthetic perfectly.
• + High-quality, realistic food photography that aligns with the specific menu items.

• − The grid is slightly asymmetrical, focusing more on a two-column split than a modular grid.

LongCat-Image

• + Dynamic use of color blocks and a more 'vibrant' accent style as requested.
• + Good variety in the food photo grid layout.

• − Text is completely illegible and garbled, failing basic professional menu standards.
• − Failed to include a specific 'Mains' section that is readable.
• − Visual artifacts and warped plate shapes in several photos.

GPT Image 1.5

• + Excellent 'exploded' effect with clearly separated, dynamic components.
• + Perfect execution of the fiery, glowing background and text effects.
• + Highly detailed food textures that look appetizing and photorealistic.

• − The crown of the 'M' in the title text is slightly cut off at the top edge.

LongCat-Image

• + Clean and legible typography for all requested text elements.
• + Good lighting on the food items making them pop against the dark background.

• − Failed the 'exploded burger' prompt; the ingredients are stacked normally rather than suspended.
• − The overall image feels more like a static graphic than a dynamic, motion-filled ad.
• − The bun texture looks slightly artificial compared to Model A.

GPT Image 1.5

• + Excellent text rendering with no spelling errors.
• + Highly realistic chalk texture with dusty smudges and natural handwriting variations.
• + Consistent and elegant cursive style throughout the entire board.

• − The background context of the 'cozy café' is barely visible compared to the close-up of the board.

LongCat-Image

• + Good environmental storytelling, showing the menu board within a nicely blurred café setting.
• + The board hardware and easel look physically realistic.

• − Significant spelling errors on almost every word (e.g., 'Toays Srays', 'Arlil').
• − Text rendering is messy and lacks the 'elegant cursive' requested in the prompt.
• − Incorrect formatting and layout compared to the prompt instructions.

GPT Image 1.5

• + Excellent cinematic lighting and texture
• + Highly detailed environment with realistic lunar dust effects
• + Superior integration of the astronaut and horse into the scene

• − The astronaut's leg positioning is slightly awkward against the horse's flank

LongCat-Image

• + Clean, bright color palette
• + Good legibility of the subject against the dark space background

• − Floating artifacts and strange aircraft shapes in the sky
• − Horse's anatomy is distorted, particularly the elongated front legs
• − Overall composition feels less cinematic and more like a collage

GPT Image 1.5

• + Excellent photorealistic texture on the capybara fur and clothing.
• + Cinematic lighting and high-quality background bokeh.
• + Perfectly captures the 'bored' expression of the passenger as requested.

• − The capybara's paws look slightly more like human fingers with hair than actual paws.
• − The taxi interior is a bit dark, making it hard to see dashboard details.

LongCat-Image

• + Clear side-view composition showing the full exterior and interior dynamic.
• + Accurately places the capybara in a professional jacket with dress shirt cuffs.
• + Good rendering of the city environment outside the window.

• − The capybara's hand/paw anatomy is distorted and looks like sharp claws.
• − The passenger is duplicated/hallucinated into two very similar people.
• − The hat is floating awkwardly above the capybara's head.

GPT Image 1.5

• + Excellent text rendering with no spelling errors in any of the requested fields.
• + Highly cohesive 'vintage gothic' aesthetic with a beautiful dark parchment texture.
• + Great composition that integrates all elements within a unified, atmospheric scene.

• − The jack-o-lantern is central but slightly smaller than the text in weight.
• − The 'parchment' effect is applied to the whole scene rather than just a poster element.

LongCat-Image

• + Good use of negative space to make the central jack-o-lantern and title stand out.
• + Vibrant colors and high contrast between the parchment and the night sky window.

• − Multiple spelling and text errors including '3uulie', '7mm', and 'The Armiees'.
• − The thorns are floating awkwardly around the paper rather than forming a natural border.
• − The composition feels fragmented with the scene appearing inside a cut-out hole in the paper.

GPT Image 1.5

• + Excellent PBR material rendering with realistic wood grain, ceramic, and liquid textures.
• + Strict adherence to the diorama base requirement and isometric perspective.
• + Clean, high-quality text rendering and professional layout.

• − Included many extra elements like a teapot and soy sauce bottle not explicitly requested in the 'minimal garnish' instruction.

LongCat-Image

• + Successfully captured the 'cartoon' and 'miniature' aesthetic with soft, clay-like textures.
• + Very clean and minimalist composition that focuses on the sushi.
• + Handled the text and small flag icon correctly.

• − The 'diorama base' is very simple and lacks the detail of a true 3D miniature scene.
• − The fish textures look more like plastic or rubber than 'realistic PBR materials' requested.

GPT Image 1.5

• + Excellent adherence to the 'tumbling together' part of the prompt with dynamic posing.
• + Superior texture rendering of fur and dew sparkles.
• + High anatomical accuracy for all four requested animals.

• − The fox kit and puppy share very similar facial structures, making them look related.

LongCat-Image

• + Strong 'god rays' and lighting effects.
• + Clear, vibrant colors in the wildflower meadow.

• − Failed to include the baby bunny, instead merging ears onto the kitten to create a 'cabbit' hybrid.
• − The animals are static and posing rather than 'playfully chasing and tumbling'.
• − The fox has an unnaturally long, stiff tail that lacks realistic physics.

GPT Image 1.5

• + Excellent vector emblem aesthetic with clean, professional typography.
• + Accurately represents the cloche dome with subtle steam and lighting.
• + High contrast and clean layout suitable for an actual logo.

• − Failed the light background requirement by using a solid black background.
• − The 'Est. 1720' text is slightly off-center within the banner.

LongCat-Image

• + Successfully incorporated the light background with subtle texture.
• + Creative integration of the text into the cloche shape.
• + Captures the warm brown and cream tones perfectly.

• − Redundant text rendering with 'Caffè' appearing twice.
• − Typographic execution is messy with overlapping letters and inconsistent line weights.
• − Composition feels cluttered compared to a professional logo.

GPT Image 1.5

• + Excellent adherence to the sequential steps requested.
• + Perfect text rendering for all mission phases.
• + Clean, professional vector aesthetic that matches the NASA-inspired palette.

• − Simple, almost clip-art style for the crew silhouettes.
• − A small typo in 'Tranquillity' (usually spelled with one 'l' in US English, though double is acceptable in other regions).

LongCat-Image

• + Creative use of a map-pin style icon for the lunar landing.
• + Follows the requested color palette reasonably well.

• − Failed significantly on text rendering with gibberish words.
• − Did not follow the specific 6-step sequence requested.
• − Icons like the rocket resemble a shuttle rather than a Saturn V, and the lunar module design is messy.