image_generation (OpenAI)
Tool : A precise action the AI can call · requires openai-api
Image generation and editing callable by an agent.
Image generation as an agent building block, with no particular attack surface.
Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.
What it does, concretely
When an agent needs to produce an image on the fly, without leaving the processing pipeline, it requires a tool that can receive a text instruction and return a usable image file directly. That is the problem image_generation solves: embedding image creation into an agent call chain, on the same level as a web search or a file read.
In practice, the tool is added to a Responses API request by declaring the image_generation type in the tools array. The model decides on its own when to trigger generation, sends the prompt to the chosen image model (for example gpt-image-2.5-sunburst for precise editing, or gpt-image-2.5-flare for speed), and returns the result as a base64-encoded image in the image_generation_call field. The text input is automatically optimized before being passed to the image model.
This tool is aimed at developers building agents or automated pipelines with the OpenAI API who need to insert an image generation step without a separate API call. It also suits teams that want fine-grained control over which image model is used depending on the use case.
It is not suited for people looking for a graphical interface to generate images manually: the tool is exclusively accessible via the API, in a programming context.
What it offers
- ·Image generation from a text prompt, triggered automatically by the agent model within a Responses API request. (paid)
- ·Support for multiple image models: gpt-image-2.5-sunburst, gpt-image-2.5-flare, gpt-image-2, gpt-image-1.5, gpt-image-1, and gpt-image-1-mini. (paid)
- ·Automatic optimization of the text input before it is passed to the image model, without manual intervention. (paid)
- ·Result returned as a base64-encoded image in the image_generation_call field, directly usable in the pipeline. (paid)
- ·Ability to provide image inputs alongside text, to guide or edit an existing image. (paid)
- ·Image model selectable at the tool level, independently of the main model used in the request. (paid)
In practice
- Price
- Paid
- Commercial use
- Commercial use: yes
- Available on
- api
- Host
- openai-api
- FR interface
- No
Security and data
- Rating
- Security: nothing to flag
- Capabilities
- readswritesexecutesnetwork
- Your data
- Data with the vendor
- Maintenance
- Active
Links and repository checked on 25 October 2026 · added on 1 April 2025
How to install image_generation (OpenAI)
tools=[{"type":"image_generation"}]