Mmx Cli

This is a Swiss Army knife for MiniMax's AI platform: text, images, video, speech synthesis with 300+ voices, and music generation all from one CLI. You'll reach for it when you need to prototype multimedia workflows or automate content pipelines without juggling multiple APIs. The speech-2.8-hd model and music-2.6 engine (lyrics, covers, instrumentals) are the standout features here. Setup is straightforward with npx or npm global install, and it includes flags like non-interactive and JSON output for agent work. If you're already committed to MiniMax or evaluating it against Stability or ElevenLabs, this gives you a single entrypoint to test everything they offer.

GitHub stars: 657
last updated: Jun 9, 2026

View on GitHub

Install

$ npx -y skills add agentscope-ai/openjudge --skill mmx-cli --agent claude-code

Installs into .claude/skills of the current project.

Files

mmx-cli Skill

Generate multimedia content (text, images, video, speech, music) via the MiniMax AI platform.

Install

npx skills add MiniMax-AI/cli@skill -g -y

Or manually:

npm install -g @minimax-ai/cli

Set your API key:

export MINIMAX_API_KEY=your_api_key_here

Capabilities

Capability	Command	Model
Text generation	`mmx text generate`	MiniMax-M3
Image generation	`mmx image generate`	image-01
Video generation	`mmx video generate`	Hailuo-2.3
Speech synthesis	`mmx speech generate`	speech-2.8-hd
Music generation	`mmx music generate`	music-2.6
Web search	`mmx search`	—

Quick Examples

# Generate text
mmx text generate "Summarize the key benefits of reinforcement learning"

# Generate an image
mmx image generate "A futuristic city skyline at sunset, photorealistic"

# Generate a short video clip
mmx video generate "A golden retriever playing in autumn leaves"

# Synthesize speech from text
mmx speech generate --text "Hello, welcome to the demo." --voice Calm_Woman

SKILL.md on GitHub