GPT Image 2 API

What is GPT Image 2 API

GPT Image 2 API provides access to OpenAI's latest image generation and editing model, which powers ChatGPT Images 2.0. It replaces DALL-E 3 and GPT Image 1.5, offering enhanced capabilities through a unified REST API.

How to use GPT Image 2 API

To use the GPT Image 2 API, you can send a POST /v2/images/generate request with your API key. The request body should include the prompt, model identifier (gpt-image-2), and any specific parameters for GPT Image 2, such as input_images, quality, background, number_of_images, output_format, and moderation. The API returns a jobId for polling or can push results to a provided webhook_url.

Features of GPT Image 2 API

• Reasoning-first generation: Includes a "thinking" mode for planning layout, web reference searching, and self-checking output.
• Batch generation: Capable of producing up to eight images from a single prompt, maintaining consistency.
• Multilingual text rendering: Accurately renders text in multiple languages and scripts.
• High resolution: Supports up to 2K output via the API.
• Flexible editing: Allows for multi-turn editing while preserving identity and context.
• Instruction following: Improved ability to follow detailed prompts and accurately place objects.
• Asynchronous jobs: Supports asynchronous job processing with webhooks for notifications.
• Reference inputs: Allows for the use of input images.

Use Cases of GPT Image 2 API

• Instruction-heavy briefs: Ideal for prompts requiring precise layout, object placement, and adherence to detailed instructions due to its reasoning capabilities.
• Posters, packaging, UI type: Suitable for designs needing legible text, including multilingual content.
• Consistent image sets: Useful for creating storyboards or campaign variants where consistency across multiple images is crucial.
• Multi-turn editing: Enables iterative design processes by maintaining identity and context across sequential edits.

Pricing

Pricing is based on credits per generation, ranging from 2 to 23 credits (approximately $0.02–$0.23) depending on the resolution and options selected. Failed jobs are refunded. A single credit balance is used across all models on Apiframe. Volume discounts are available.

FAQ

• What is GPT Image 2? OpenAI's newest image generation and editing model, powering ChatGPT Images 2.0 and replacing DALL-E 3 and GPT Image 1.5.
• What is "thinking" mode? A reasoning step where the model plans layout, searches for references, and reviews its output before generation, enabling up to eight images per prompt.
• How is it different from GPT Image 1.5? It features built-in reasoning, stronger multilingual text capabilities, consistent multi-image sets, and higher overall quality.
• Can it render text in multiple languages? Yes, multilingual text rendering across several scripts is a key strength.
• Is it free? Standard mode is free for ChatGPT users. API access is paid on a token basis.
• Where can you access it? Through Apiframe, ChatGPT, and the OpenAI API.