GPT Image by OpenAI

GPT Image 1 is an image model focused on image editing and transformation rather than pure text understanding. It performs best when given rich, narrative-style prompts that clearly describe the desired visual outcome, context, style, and details, instead of short or command-like instructions.

Use this model

Speed

Price

Inputs

TextImage

Outputs

Image

Lab

OpenAI

GPT Image by OpenAI sample outputs

Captured directly from our eval suite. Click any tile to inspect the full render.

Realistic portrait test

“A kodachrome 25 photograph of a woman with curly brown hair walking through a busy city street, natural lighting, dynamic composition, 35mm lens, authentic urban atmosphere”

EngineId: gpt-image-1

•

Quality: auto

•

Image Size: square_hd

Text rendering capability test

“A vintage poster with the text 'GLIF ALLHANDS 2025' in bold retro typography, with undertext saying 'Reykjavik, Iceland', vibrant colors, psychedelic style with glossy 3d elements”

EngineId: gpt-image-1

•

Quality: auto

•

Image Size: square_hd

Testing very long and detailed prompts

“studio photograph of a silver panther on the left, a giant beach ball in the middle and a tall man on the right. the panther has long fur and green horns, its claws are attached to garden hoses that end up in milkshakes, the left scene is a highway restaurant. The beachball has chrome and black sections and is in surrounded by water, the man is wearing hi-viz street fashion and is cosplaying as a techbro and is surrounded by multiple signs that 'we're so back'”

EngineId: gpt-image-1

•

Quality: auto

•

Image Size: square_hd

Vague prompt test

“determinism”

EngineId: gpt-image-1

•

Quality: auto

•

Image Size: square_hd

Social media selfie test

“male power tool influencer taking a selfie, perfect lighting, trendy outfit, confident smile, modern background, candid raw iPhone photo style”

EngineId: gpt-image-1

•

Quality: auto

•

Image Size: square_hd

Check if a model can do miniature scenes inherently

“Miniature tilt-shift scene of a wooden trebuchet siege engine in a medieval military warcamp, with canvas tents scattered around, armored soldiers and engineers working on the catapult, wooden fortifications in background, campfires smoking, banners fluttering in the wind, muddy ground with wagon tracks, medieval era setting”

EngineId: gpt-image-1

•

Quality: auto

•

Image Size: square_hd

Suprise prompt

“grindle flompa zeet varoo, plingle drask in bloofer haze — twomple zarn reega 7, snibble dront kalee swoop, mirra gongle thru floovian glass, blorpa tect frindle noom, abstract glibber fog machine of gloon and jell, high surreal detail, untranslatable energy field, visual nonsense in motion”

EngineId: gpt-image-1

•

Quality: auto

•

Image Size: square_hd

Test if model can stack animals in the right order on top of each other

“studio photo a group of stacked animals: a rooster sitting on the back fo a cat which is sitting on the back of a dog which is sitting on the back of a donkey, riding through a small german medieval town, shot in the 1980s”

EngineId: gpt-image-1

•

Quality: auto

•

Image Size: square_hd

Magic laptop test

“An ancient-looking laptop covered in mystical runes, its screen displaying swirling galaxies, used to hack into magical networks or cast digital spells.”

EngineId: gpt-image-1

•

Quality: auto

•

Image Size: square_hd

TCG card test

“Create a trading card of a anime mecha-hamster with elite coding skills”

EngineId: gpt-image-1

•

Quality: auto

•

Image Size: square_hd

Style transfer test

“Transform into the style of a vintage rave poster”

EngineId: gpt-image-1

•

Quality: auto

•

Image Size: square_hd

Edit parts of image test

“Open up the cyborg's right part of their head and show the underlying electronics”

EngineId: gpt-image-1

•

Quality: auto

•

Image Size: square_hd

Launch your first agent today

Build without code. Share without limits.

Create an Agent