OpenAI has expanded its innovative image generation capabilities, originally introduced in ChatGPT, by integrating the technology into its API. This move enables developers to incorporate these advanced features into their own applications and services. Launched for most ChatGPT users in March, the image generator quickly gained popularity, particularly for its ability to create striking Ghibli-style images and “AI action figures.” This surge in usage resulted in over 130 million ChatGPT users generating more than 700 million images within the first week, consequently putting significant pressure on OpenAI’s infrastructure.
The engine behind this image generation is called “gpt-image-1,” a multimodal AI model capable of producing images in various styles while adhering to custom specifications, drawing on extensive world knowledge, and even incorporating text. Developers have the flexibility to generate multiple images simultaneously using gpt-image-1, allowing them to manage both the quality and speed of the output effectively.
OpenAI maintains a robust set of safety protocols within gpt-image-1, similar to those in ChatGPT. These guidelines prevent the generation of content that violates the company’s policies. Developers have the option to adjust moderation sensitivity settings, with choices ranging from standard filtering to a less restricted mode, enabling greater control over potentially age-inappropriate content.
Additionally, all images produced using gpt-image-1 are watermarked with C2PA metadata, ensuring they can be identified as AI-generated by compatible platforms.
In terms of pricing, developers are charged $5 per million input tokens for text, $10 for images, and $40 for each million output tokens for images. This translates to approximately 2 cents for low-quality images, 7 cents for medium, and 19 cents for high-quality square images.
Several major companies, such as Adobe, Airtable, Wix, and Canva, are experimenting with or actively implementing gpt-image-1 in their services. Notably, Figma’s design platform now allows users to generate and modify images using this advanced model, while Instacart is looking into the potential for creating images related to recipes and shopping lists. OpenAI’s latest development highlights its commitment to enhancing creative tools in the digital landscape through advanced AI technology.
Fanpage: TechArena.au
Watch more about AI – Artificial Intelligence
