All products

Create images and audio with MiniMax AI

Use the official MiniMax command-line tools for image generation, speech and transcription, and track longer video requests without losing the job reference.

Teams can create images and audio with MiniMax AI from a single agent environment while keeping each request separate. The minimax-multimodal-toolkit skill explains how to choose a command, preview the payload and save the output in the intended place.

For a short training module, start with a cover illustration and a few lines of narration. Listening to that sample before generating the complete script makes it easier to catch pronunciation issues. A later video request can use an approved still and return a job ID while it runs.

What it does

  • Generate an image from an original text brief.
  • Create spoken audio using an authorised voice selection.
  • Transcribe an audio file that you are permitted to send to the service.
  • Preview a request before submitting a billable operation.
  • Track an asynchronous video task and retrieve its completed output.

Install and use

  1. Install the skill and official MiniMax CLI, then configure an authorised account through a private credential source. Features and charges depend on the selected model.
  2. Ask “Prepare a cover image for this lesson. Show me the request before you generate it and save the result in the lesson folder.”
  3. Use a short narration sample for the audio step and review it before requesting the full recording.

FAQ

Source

Made by MiniMax. Original skill: minimax-multimodal-toolkit, published on GitHub under the MIT licence.

View on GitHub