FruxonDocs

OpenAI Media

Generate images and transcribe audio using OpenAI models (GPT-Image, DALL-E, Whisper)

The OpenAI Media integration lets your agents generate images and transcribe audio using OpenAI's models.

Prerequisites

You need:

Setup

  1. Open your agent in Agent Studio
  2. In the Integrations panel, click Add Integration Config
  3. Select OpenAI Media from the integration list
  4. Give the config a display name (e.g., "OpenAI Media")
  5. Paste your API Key
  6. Save the agent revision

Using in Your Agent

  1. In an Agent Step, attach OpenAI Media tools from the tools panel
  2. The agent uses your API key for all OpenAI Media API calls
  3. Tools are referenced as openai_media.gpt-image-1, etc.

Available Tools

Image Generation

ToolDescription
gpt-image-1Generate images with GPT Image 1
gpt-image-1-miniGenerate images with GPT Image 1 Mini (faster, lower cost)
gpt-image-1.5Generate images with GPT Image 1.5

Audio

ToolDescription
transcribe_audioTranscribe audio to text using Whisper

On this page