OpenAI · Photo

GPT Image 1.5 Text-to-Image

Images from textRealistic imagesArtProduct shots

GPT Image 1.5 (T2I) is OpenAI’s fast, cost-efficient text-to-image generator powered by GPT-5 guidance. Create photorealistic shots, product renders, concept art, and stylized graphics from natural-language prompts (optionally conditioned with an image).

from€0.08per render
Use in Genlaxy →

Examples

GPT Image 1.5 Text-to-Image: Street food market in Tokyo at night, chef tossing flaming wok with vegetables mid-air, steam rising, colorful
Prompt

Street food market in Tokyo at night, chef tossing flaming wok with vegetables mid-air, steam rising, colorful paper lanterns overhead, motion blur on crowd in background, vibrant neon signs, photojournalism style

Try this prompt →
GPT Image 1.5 Text-to-Image: Portrait of a cyberpunk hacker, neon pink and blue rim lighting, chrome cybernetic implants on temple, rain dr
Prompt

Portrait of a cyberpunk hacker, neon pink and blue rim lighting, chrome cybernetic implants on temple, rain droplets on skin, dark alley background with holographic advertisements, hyper-detailed, 8K

Try this prompt →
GPT Image 1.5 Text-to-Image: Macro photograph of morning dew drops on spider web, each droplet reflecting a tiny forest landscape, soft bok
Prompt

Macro photograph of morning dew drops on spider web, each droplet reflecting a tiny forest landscape, soft bokeh background, golden backlight, extreme detail, nature photography

Try this prompt →
GPT Image 1.5 Text-to-Image: Surrealist artwork of a giant whale swimming through clouds above a tiny village, bioluminescent patterns on w
Prompt

Surrealist artwork of a giant whale swimming through clouds above a tiny village, bioluminescent patterns on whale skin, villagers looking up in wonder, dreamlike atmosphere, style of Salvador Dalí meets Hayao Miyazaki

Try this prompt →

Specs

Typetext-to-image
OutputPhoto
InputText
Model IDopenai/gpt-image-1.5/text-to-image

Similar models

All models →

Prices are per render and charged from your balance only when a render succeeds. Failed renders are refunded automatically. Examples show what the model can do; your result depends on your prompt and settings.