Which providers does this MCP server support?

OpenAI, xAI, Google Gemini, ElevenLabs, and BFL (Flux). Each provider supports different capabilities — OpenAI and Gemini cover images/video/audio/transcription, while BFL specializes in FLUX image generation and editing.

How do I set up authentication for multiple providers?

Set the corresponding API key environment variables: OPENAI_API_KEY, XAI_API_KEY, GEMINI_API_KEY, ELEVENLABS_API_KEY, and BFL_API_KEY. You only need to configure the providers you plan to use.

Where are generated media files saved?

By default, media files are saved to the current working directory. You can customize the output location by setting the MEDIA_OUTPUT_DIR environment variable.

Can I switch between providers for the same task?

Yes — the unified interface lets you call different providers for the same capability (e.g., image generation via OpenAI, xAI, Gemini, or BFL) without changing your code.

What's the difference between this and calling each API directly?

This MCP server abstracts provider-specific request formats and authentication into a single interface, reducing boilerplate and making it easier to swap providers or use multiple providers in the same application.

MCP ServerSTDIOOfficialv1.3.1

Rsmdt Multimodal MCP Server

Generate images, video, audio, and transcriptions across OpenAI, xAI, Google Gemini, ElevenLabs, and BFL providers through a single unified interface. For developers building multi-modal AI applications who want provider flexibility without rewriting integration code.

io.github.rsmdt/multimodal

Try MCP with AI models View source

Hosted URL

Local install

Transport

STDIO

Auth

No auth required

What the Rsmdt Multimodal MCP server does

How models use it and what it is built for.

Connect to Rsmdt Multimodal

Local install — runs as a subprocess.

npx @r16t/multimodal-mcp@1.3.1

Environment variables

Configuration this server reads at startup.

OPENAI_API_KEYSecret
OpenAI API key for image, video, audio generation and transcription
XAI_API_KEYSecret
xAI API key for image and video generation
GEMINI_API_KEYSecret
Google Gemini API key for image, video, and audio generation
ELEVENLABS_API_KEYSecret
ElevenLabs API key for audio generation and transcription
BFL_API_KEYSecret
BFL API key for FLUX image generation and editing
MEDIA_OUTPUT_DIR
Directory for saved media files (defaults to cwd)

Resources

Where to find authoritative docs and source for Rsmdt Multimodal.

Source repository

Example prompts for Rsmdt Multimodal

Paste any of these into Agent Studio after connecting Rsmdt Multimodal.

Generate a 1024x1024 image of a mountain landscape using FLUX
Transcribe this audio file and return the text
Create a 10-second video clip of a sunset with OpenAI
Generate a natural-sounding voiceover in English using ElevenLabs

Rsmdt Multimodal MCP server — FAQ

Common questions about connecting and running Rsmdt Multimodal.

Which providers does this MCP server support?
OpenAI, xAI, Google Gemini, ElevenLabs, and BFL (Flux). Each provider supports different capabilities — OpenAI and Gemini cover images/video/audio/transcription, while BFL specializes in FLUX image generation and editing.
How do I set up authentication for multiple providers?
Set the corresponding API key environment variables: OPENAI_API_KEY, XAI_API_KEY, GEMINI_API_KEY, ELEVENLABS_API_KEY, and BFL_API_KEY. You only need to configure the providers you plan to use.
Where are generated media files saved?
By default, media files are saved to the current working directory. You can customize the output location by setting the MEDIA_OUTPUT_DIR environment variable.
Can I switch between providers for the same task?
Yes — the unified interface lets you call different providers for the same capability (e.g., image generation via OpenAI, xAI, Gemini, or BFL) without changing your code.
What's the difference between this and calling each API directly?
This MCP server abstracts provider-specific request formats and authentication into a single interface, reducing boilerplate and making it easier to swap providers or use multiple providers in the same application.

Skip the local setup — run MCP in your browser

MCP Playground runs 10,000+ hosted MCP servers — GitHub, Linear, Notion, Stripe, Sentry and more — across Claude, GPT, Gemini, DeepSeek and 60+ AI models. Compare model answers side-by-side, save agent presets, share runs. Zero install.

Open Agent Studio