Jimeng MCP Server

Example Use Cases

• 1

  Generate high-quality images from text prompts using multiple Jimeng models and receive direct URLs for display.

• 2

  Produce videos (with optional first/last frame images) using Jimeng video models and obtain playable video URLs.

• 3

  Perform image-mixing or reference-based generation by providing local or remote images as inputs and automatically uploading them to Jimeng.

Description

This project implements a Model Context Protocol (MCP) server in TypeScript that directly calls Jimeng AI's image and video generation APIs (via reverse-engineered endpoints). It supports multiple Jimeng models, image mixing/reference generation (local or remote images), video generation with optional first/last frame customization, and various image/video parameters. The server can run as a stdio MCP for agents like Claude Desktop or as an HTTP API service, and automates uploading and handling of reference images.