MiniMax released an official open-source CLI tool in April 2026 that puts AI video, music, image, and speech generation directly into the terminal. Version 1.0.7 of the tool ships seven distinct capabilities under one command surface, giving developers and power users a single interface to access the MiniMax API without leaving their workflow.

What Happened

The MiniMax CLI, launched on April 10 and featured on Product Hunt the same day, is an MIT-licensed TypeScript tool installable via npm. Running npm install -g mmx-cli on any Node.js 18+ system gives immediate access to seven modes:

  • Text: Multi-turn chat with streaming, system prompts, and JSON output
  • Image: Text-to-image generation with aspect ratio and batch controls
  • Video: Asynchronous video generation with progress tracking
  • Speech: Text-to-speech with 30-plus voices, speed control, and streaming playback
  • Music: Text-to-music including lyrics mode, instrumental mode, and cover generation from a reference audio file
  • Vision: Image understanding and descriptive analysis
  • Search: Web search from the command line

Core commands follow a consistent pattern: mmx text chat, mmx image generate, mmx video generate, mmx speech synthesize, mmx music generate. The tool can also be loaded as an AI agent skill with npx skills add MiniMax-AI/cli -y -g. API access requires a MiniMax Token Plan.

Why It Matters for Creators

Most multimodal AI tools require switching between browser tabs, uploading files through web forms, and waiting for dashboards to load. The MiniMax CLI eliminates that overhead. A creator can generate a voiceover, produce a background music track, and render a short video clip in a single terminal session, then pipe results directly into an existing build or post-processing script.

The music capability stands out in particular. Generating an original instrumental track or a full song with lyrics from a text prompt, and creating audio covers from reference recordings, are tasks that typically require dedicated apps. Having them available as a single mmx music generate command with no UI dependency is a meaningful shift for audio-forward workflows.

The MIT license means the CLI can be embedded into automation pipelines, CI systems, or custom tooling without licensing concerns. With 131 commits already on the repository, the codebase is not a proof of concept.

What to Do Next

Install with npm install -g mmx-cli (requires Node.js 18 or higher). Review the GitHub repository for full command documentation and configuration options. Obtain a MiniMax Token Plan API key before running any generation commands. Test mmx music generate with an instrumental prompt to evaluate audio quality for your workflow.


This story was covered by Creative AI News.

Subscribe for free to get the weekly digest every Tuesday.