OpenAI's latest advanced image generation model is now available through developers API

OpenAI has made its latest advanced image generation model available to developers through the OpenAI's API, powered by an AI model called “gpt-image-1”. Initially launched last month in ChatGPT, saw significant engagement with over 700 million images created by 130 million users in just a week, largely driven by the "Ghibli style" trend. The API allows developers to generate high-quality images with diverse styles and precise editing capabilities, including accurate text rendering.

Developers can adjust moderation sensitivity and configure settings such as quality, quantity, background transparency, and output formats. The usage-based pricing is set at $5 per 1M tokens for text input, $10 per 1M for image input, and $40 per 1M for image output, with typical per-image costs ranging from $0.02 to $0.19 depending on quality.

gpt-image-1 is set to be incorporated into the Responses API, broadening its reach beyond the current Images API. Developers can explore the model in the OpenAI Playground. Companies like Adobe Firefly, Adobe Express, Wix.com, Microsoft Azure, Airtable, Figma, Instacart or Canva are already integrating this technology.