GPT Image 1 OpenAI vs LongCat-Image Meituan

GPT Image 1

• + Excellent adherence to lighting instructions (soft light from the left)
• + Clean and realistic book texture
• + Good plant placement with clear visibility through glass

• − The sphere appears to be floating unnaturally above the base
• − The blue sphere has a matte texture which was unspecified but looks slightly less like a classic 'sphere' object than Model B

LongCat-Image

• + Realistic physics with the sphere resting on the bottom surface
• + Beautiful light refraction and reflection on the glass cube and sphere
• + Strong prompt adherence including the window background

• − The plant is slightly less 'behind' the cube than in Model A, appearing more to the side

GPT Image 1

• + Excellent skin texture and realistic portrayal of an elderly person.
• + Accurate shallow depth of field that feels professional and cinematic.
• + Higher quality rendering of textures like wet fabric and metal.

• − Missed the request for motion blur on passing cars.
• − Framing is very tight, losing some of the street atmosphere.

LongCat-Image

• + Successfully captured motion blur from passing cars.
• + Good use of reflections on the wet pavement.
• + Includes visible rain streaks that add to the atmosphere.

• − Anatomical and structural issues with the bicycle (multiple wheels/frames merging).
• − Lower visual quality with a slightly plastic, AI-generated look to the faces.
• − The rain looks somewhat like a filter rather than a natural part of the scene.

GPT Image 1

• + Excellent skin texture and realistic, battle-worn appearance with convincing dirt and subtle scars.
• + The lighting and bokeh sparks feel integrated and atmospheric.
• + Ornate engraving on the armor is incredibly detailed with realistic weathering.

• − The beads in the hair are very subtle and blend in too much with the hair color.
• − Composition is a bit tight, losing some of the promised 'leather straps and cloth underlayer' details.

LongCat-Image

• + Very clear interpretation of the beads in the hair with vibrant colors.
• + Excellent visibility and detail on the leather straps, buckles, and chainmail/cloth transition.
• + Strong cinematic lighting with clear torchlight reflection.

• − The scars look a bit 'painted on' rather than being part of the skin texture.
• − The character looks a bit too pristine and 'pretty' for a battle-worn description.
• − The sparks in the top right are a bit distracting and look like digital artifacts.

GPT Image 1

• + Strong text legibility with clean, professional sans-serif fonts.
• + Consistent high-quality food photography that fits a professional menu style.
• + Strict adherence to the 2-column minimalist layout requested.

• − Repetitive placeholder text with spelling errors ('Apperoiation descrigion').
• − Limited variety in the grid layout, focusing only on two main columns.

LongCat-Image

• + Dynamic and colorful layout with vibrant blue and yellow accents.
• + Includes a larger variety of food imagery across the design.

• − Text is illegible and contains severe AI artifacts/gibberish.
• − Layout feels cluttered and does not follow a clean professional grid as well as the competitor.
• − Source imagery has some warped elements and inconsistent lighting.

GPT Image 1

• + Excellent adherence to the 'exploded' layout with all components suspended separately.
• + Consistent fiery, glowing effect applied to all text elements as requested.
• + High level of texture detail on the bun, tomato, and patty.

• − The price text is missing the number '6', displaying only '.99'.
• − The starburst shape for the price is slightly irregular.

LongCat-Image

• + Accurate rendering of the numerical price and secondary message within the starburst.
• + Dynamic background featuring literal coals and flames that enhances the theme.
• + Glassy, high-quality lighting on the main title text.

• − Failed the 'exploded' layout requirement, as the burger is mostly assembled rather than suspended in mid-air.
• − The secondary message is integrated into the starburst rather than being a separate element as implied by the layout.

GPT Image 1

• + Excellent text legibility and spelling accuracy.
• + The chalk texture is highly realistic with believable stroke variations.
• + The composition is clean and centered, making it easy to read.

• − The font lacks an 'elegant cursive' style as requested, appearing more like standard print.
• − The 'cozy café' background is missing, showing only the board.

LongCat-Image

• + Successfully captured the 'cozy café' setting in the background.
• + The chalk handwriting has a more artistic, hand-drawn flair to it.

• − Significant spelling errors throughout almost every word.
• − Failures in text rendering, with characters blending into illegible scribbles.
• − The layout is cluttered and the pricing does not align correctly with items.

GPT Image 1

• + Excellent cinematic lighting and dark atmosphere
• + Anatomically coherent horse and astronaut interaction
• + Clean, high-resolution rendering with realistic textures

• − Fails the specific negative constraint to have the horse on top of the astronaut

LongCat-Image

• + Includes interesting background elements like space debris and planets
• + Vibrantly colored assets with a clear lunar-style landscape

• − Fails the specific negative constraint to have the horse on top of the astronaut
• − Messy composition with floating artifacts and incoherent background objects
• − Anatomical issues with the horse's back legs and the astronaut's scale

GPT Image 1

• + Excellent photorealistic texture on the capybara's fur
• + Cinematic lighting and high-quality bokeh in the background
• + Captures the bored expression of the passenger perfectly

• − The capybara's paws look more like human-monkey hybrid hands than rodent paws
• − The viewpoint is from the front hood looking in, rather than a side profile of the interior

LongCat-Image

• + Shows more of the taxi exterior and environment for context
• + Large, detailed capybara with accurate-looking fur and ears
• + Includes the uniform jacket with a dress shirt as requested

• − Anatomical failure with the passenger, featuring two faces for one body
• − The capybara's hand/claw is poorly rendered and not properly gripping the wheel
• − The hat is floating weirdly on the head rather than being worn

GPT Image 1

• + Perfect text accuracy for all requested fields including the date and location.
• + Sophisticated vintage aesthetic with atmospheric lighting.
• + Clean, professional layout that matches the gothic invitation theme.

• − The scrolls and banner elements are a bit subtle compared to the dark background.

LongCat-Image

• + Creative use of a torn parchment effect and physical thorn borders.
• + Stronger 'gothic' typeface for the main header.

• − Significant spelling errors in the event details such as 'The Armiees' and '7umm'.
• − Layout feels cluttered with inconsistent bat styles and jagged cutouts.
• − Failed to place the small banner text in a scroll as requested.

GPT Image 1

• + Excellent typography rendering with clean, centered alignment.
• + Consistent soft-clay 3D aesthetic that matches the 'cartoon scene' prompt perfectly.
• + High-quality lighting and shadows that give the diorama realistic depth.

• − The chopsticks are slightly merged into the diorama base.
• − The fish texture is somewhat uniform and lacks the detail seen in organic materials.

LongCat-Image

• + Beautiful subsurface scattering and textures on the salmon and tuna pieces.
• + Accurate interpretation of a traditional wooden sushi plinth as the diorama base.
• + Dynamic Flag icon with a waving effect.

• − The text alignment is slightly off-center and tilted.
• − The 'JAPAN' text has a slight visual artifact on the letter 'N'.
• − The scale of the garnish (wasabi) is a bit small compared to the sushi.

GPT Image 1

• + Excellent adherence to the prompt with all four distinct animals clearly represented.
• + Very convincing interaction and dynamic 'tumbling' motion between the animals.
• + Subtle and natural-looking golden hour lighting with realistic god rays.

• − The fox kit has dark front paws that almost look like detached silhouettes in the grass.

LongCat-Image

• + Bright and colorful composition with prominent butterflies.
• + Strong 'god ray' effect coming from the top left corner.

• − Failed to include a separate baby bunny, instead merging features to create a cat with rabbit ears.
• − Anatomical issues including the dog having five legs/paws visible in the gait.
• − The dew sparkles look like floating glass orbs rather than natural moisture.

GPT Image 1

• + Clean vector illustration style
• + Excellent typography and accurate spelling
• + Strong adherence to the minimalist aspect of the prompt

• − Solid black background ignores the 'light background' request
• − Slightly too simplified for a vintage style

LongCat-Image

• + Successfully uses a light background with subtle texture
• + Rich illustrative style that feels more authentically 'vintage'
• + Dynamic steam and banner design

• − Significant spelling errors with 'Caffè' appearing twice and incorrectly merged
• − Cluttered composition compared to a true logo
• − Line art on the cloche is slightly messy

GPT Image 1

• + Excellent typography with mostly legible text such as 'ARMSTRONG' and 'TRANQUILITY'.
• + Followed the color palette and flat-vector style request very well.
• + Includes most mission steps logically across the layout.

• − Has one significant spelling error for the step 'EARLLUNAR' and skips 'Lunar Orbit' as its own labeled step.
• − The layout is a bit cluttered with mismatched labels and icons.

LongCat-Image

• + Features a very clean and professional layout with good use of negative space.
• + Iconography is sharp and consistent across the design.
• + Excellent interpretation of the 'NASA-inspired' palette.

• − Failed significantly on text rendering with gibberish words like 'Aalneiitn'.
• − Missed the specific sequential mission steps requested in the prompt.
• − The rocket icon looks more like a shuttle than a Saturn V.