​​OpenAI Launches GPT-Image-1 Model via API

OpenAI has unveiled its latest advancement in artificial intelligence with the launch of the GPT-Image-1 API, a powerful image generation model now accessible to developers and businesses. Building upon the capabilities of GPT-4o, GPT-Image-1 offers enhanced versatility in creating images across diverse styles, accurately rendering text, and incorporating world knowledge into visual outputs.

Adobe has integrated GPT-Image-1 into its Firefly and Express applications, providing users with expanded creative flexibility. This integration allows for the generation of images in various aesthetic styles, catering to the needs of business professionals, consumers, and creators alike. Figma has also adopted the model, enabling users to generate and edit images directly within its design platform through simple prompts, streamlining the design process and fostering rapid ideation.

In addition to Adobe and Figma, other companies such as Canva, GoDaddy, and Instacart are exploring applications of GPT-Image-1, signaling a broader industry shift towards AI-driven image generation. The model is currently available through OpenAI’s Images API, with plans to extend support to the Responses API in the near future.

The introduction of GPT-Image-1 follows OpenAI’s recent $40 billion funding round, led by SoftBank, which has bolstered the company’s valuation to $300 billion. This financial backing underscores the growing confidence in OpenAI’s AI innovations and their potential to transform various sectors.

As AI-generated imagery becomes increasingly prevalent, considerations regarding ethical use and environmental impact have come to the forefront. OpenAI has implemented restrictions on generating images in the style of certain living artists to address copyright concerns. Additionally, the energy demands of AI models like GPT-Image-1 have prompted discussions about sustainability and the need for more efficient data center operations.

With GPT-Image-1, OpenAI continues to push the boundaries of what’s possible in AI-driven creativity, offering tools that empower users to bring their visual ideas to life with unprecedented ease and precision.

Pricing Structure of GPT-Image-1

The API employs a token-based pricing model:

• Text Input Tokens: $5 per 1 million tokens.
• Image Input Tokens: $10 per 1 million tokens.
• Image Output Tokens: $40 per 1 million tokens.​

In practical terms, this translates to approximately:

• Low-Quality Images: $0.02 per image.
• Medium-Quality Images: $0.07 per image.
• High-Quality Images: $0.19 per image.

Getting Started

Developers can access GPT-image-1 API API through CometAPI. To begin, explore the model’s capabilities in the Playground and consult the API guide (model name: gpt-image-1) for detailed instructions. Note that some developers may need to verify their organization before using the model.