Built by the NLB team
Every media model, one command line
Generate images, video, speech and music, transcribe and analyze media, and convert documents to Markdown from a single provider-aware CLI built for people and AI agents.
Read the docs(opens in a new tab)Gallery
Features
Generate
Image, video, speech and music generation across 12 AI providers.
Analyze and transcribe
Media analysis and transcription, for example Gemini analyze and transcribe, and speech to text.
Documents to Markdown
Convert documents into clean Markdown with multix doc convert.
Optimize media
Resize and compress with ffmpeg and ImageMagick, including a target file size.
Key rotation
Gemini API keys rotate automatically when a rate limit is hit.
Self-checking
multix check validates your setup and multix update upgrades in place.
Technology stack
- Runtime
- Node.js 20+
- Language
- TypeScript
- CLI
- commander
- Validation
- zod
- Media
- ffmpeg
- ImageMagick 7
- Docs site
- Astro Starlight
Pricing
Open source
PopularFree/MIT license
- Every command and provider adapter
- Bring your own provider keys
- Markdown docs and llms.txt for agents
Questions
How do I install multix?
Run npm install -g @mrgoonie/multix (Node.js 20 or newer), then multix check to validate your setup.
Which providers does it support?
OpenAI, Gemini, MiniMax, OpenRouter, Leonardo.Ai, BytePlus, Cloudflare, ElevenLabs, fal.ai, Atlas Cloud, Replicate and HeyGen.
Is it open source?
Yes. multix is MIT licensed and developed in the open at github.com/mrgoonie/multix-cli.
Maker
Duy Nguyen
CTO and co-founder of T.O.P Group, creator of GoClaw and UI UX Pro Max, and founder of the Build in Public Vietnam community.
Give your agents a media toolbox
Install multix and generate, analyze or convert media from one command.
Star it on GitHub(opens in a new tab)