OpenAI launches a new generative model for creating images via API

OpenAI has introduced a new generative model for image creation, available via API — gpt-image-1. The model enables developers and businesses to integrate high-quality image generation into their tools and platforms. The new model offers flexibility and high accuracy, generating images in various styles while precisely following user instructions, leveraging world knowledge, and correctly handling text.

Many leading companies, including Adobe, Airtable, Figma, Wix, and others, have already integrated gpt-image-1 capabilities into their products, enhancing functionality for image generation and editing, creating marketing materials, and improving visual effects.

Adobe is leveraging these new capabilities to integrate image generation into its Firefly and Express applications, providing users with flexibility to experiment with different styles and improving creative production workflows. Figma uses the model for generating and editing images, allowing designers to quickly bring ideas to life and enhance visual concepts. Wix has integrated the technology into its Wixel platform, enabling users to create professional designs with minimal effort.

The model also provides advanced capabilities for enhancing avatars, creating personalized YouTube thumbnails, improving product quality for online sales, and many other applications. The pricing for using gpt-image-1 is based on the number of tokens, with separate pricing for text tokens, image tokens, and generated images.

Source: openai.com