by n24q02m
Imagine MCP is an MCP server that lets AI agents understand images and videos and generate new ones through Gemini, OpenAI and Grok. The repository is archived and no longer maintained.
PluginContact for pricing
Step 1: Get an API key from at least one supported provider: Gemini, OpenAI or xAI.
Step 2: Run the server with uvx imagine-mcp, or pull the Docker image from ghcr.io.
Step 3: Add an imagine entry to your MCP client config with the command uvx, the argument imagine-mcp and your key in env.
Step 4: Optionally set UNDERSTAND_MODELS or GENERATE_MODELS to choose models and fallback order.
Step 5: Ask your agent to understand or generate an image or video, passing a provider and tier if you want to override the default.
No. The README carries an archived notice dated 2026-09-13 saying the repository is not maintained. Existing installations keep working but get no updates or support.
Gemini, OpenAI and Grok for generation, with understanding routed through litellm so other litellm providers also work.
It starts without any, but in stdio mode it exits if none of the three provider keys are set. Add at least one key to use it.
Requirements
Vuela MCPVuela MCP connects Vuela's content tools to Claude, ChatGPT and other MCP assistants. You ask in plain language and videos, images, voiceovers and SEO articles are created in your Vuela account.
Morpha MCPMorpha MCP lets an AI assistant build and edit short-form videos in Morpha, a browser video editor for Reels, TikTok and Shorts. You then hand-tune the result in the editor and render it.
Apatero Studio MCPApatero Studio MCP connects Claude to Apatero's hosted creative studio. You can generate images and video, keep persistent AI characters and build batches of posts from chat.