GPT Image 2
GPT Image 2 is an image generation model from OpenAI that creates images from text prompts and supports source image inputs.
Text-to-image generation with source image input
GPT Image 2 is an image generation model developed by OpenAI, released in April 2026. It accepts text prompts alongside optional source images, allowing users to generate new images informed by existing visual references. The model supports configurable output size and format options, giving developers control over the shape and file type of generated results.
GPT Image 2 is suited for tasks that involve creating or transforming images based on textual descriptions, including content creation, design prototyping, and visual ideation workflows. Pricing is usage-based, ranging from approximately $0.05 to $0.20 per image depending on configuration. The model is available directly through MindStudio without requiring separate API key setup.
What GPT Image 2 supports
Text-to-Image Generation
Generates images from natural language text prompts using a 4,000-token context window for prompt input.
Source Image Input
Accepts one or more source images as visual references via an image URL array input, enabling image-informed generation.
Configurable Output Size
Supports selectable image size options so developers can specify dimensions appropriate for their use case.
Output Format Selection
Allows users to choose the output file format for generated images, giving control over how results are delivered.
Usage-Based Pricing
Billed per image generated at approximately $0.05–$0.20 per image, with cost varying by configuration.
Ready to build with GPT Image 2?
Get Started FreeCommon questions about GPT Image 2
What is the context window for GPT Image 2?
GPT Image 2 has a context window of 4,000 tokens, which applies to the text prompt input used to guide image generation.
How much does GPT Image 2 cost to use?
Pricing is approximately $0.05 to $0.20 per image generated, with the exact cost depending on the selected size and output format configuration.
Can I provide existing images as input to GPT Image 2?
Yes. GPT Image 2 accepts source images via an image URL array input, allowing you to supply visual references that inform the generated output.
What output formats and sizes does GPT Image 2 support?
The model includes selectable options for both output size and output format. The specific available values are configurable through the model's input parameters in MindStudio.
When was GPT Image 2 released?
GPT Image 2 was released in April 2026 and is published by OpenAI.
Parameters & options
If you want to edit an existing image, provide the URL(s) or variables
Explore similar models
Start building with GPT Image 2
No API keys required. Create AI-powered workflows with GPT Image 2 in minutes — free.