ImgMCP is a multimedia AI model aggregation platform. It aims to provide you with a concise and efficient way to access and utilize various effective multimedia AI models.
We offer a unified API and the MCP (Multimedia Control Protocol) to simplify the model integration process—whether you are integrating them directly into your applications or invoking them via a Large Language Model (LLM).
Our goal is to connect the capabilities of these AI models with your creativity, striving to enhance the experience, reduce costs, and increase efficiency for every creator.
Please consult the relevant documentation here to learn how to leverage ImgMCP's capabilities.
Before you start using the ImgMCP platform, please ensure you have the following ready:
An ImgMCP Account: You need to register on the platform. If you don't have an account yet, please go to the Dashboard to register.
An API Key: An API key is required to authenticate your requests. After registering and logging in, you can find your default API key on the API Keys page in the management console.
This section will guide you on how to configure an MCP Host (such as Cherry Studio) to interact with the ImgMCP platform via the provided Streamable HTTP MCP Service. This method typically doesn't require installing additional client libraries; you only need to set up the service endpoint and API key.
Supported MCP Hosts:
Native Support: Tools like Cherry Studio and ChatWise offer a good integration experience.
Via mcp-remote: For Hosts that do not yet support the Streamable HTTP protocol (like Claude Desktop), you can use the mcp-remote tool in conjunction.
(For detailed tool configuration methods, please refer to our dedicated MCP documentation. The following steps use Cherry Studio as an example.)
1. Service Endpoint & Authentication:
MCP Endpoint URL:
https://api.imgmcp.com/mcp
Authentication Method: ImgMCP uses API keys for authentication. You need to provide the key in the Authorization header of your HTTP requests in the following format:
Authorization: Bearer <Your API Key>
Tip: If configuring the Bearer <Your API Key> format is inconvenient in your MCP Host, you can try using your API key directly as the value for the Authorization header. However, using the standard format including Bearer is recommended whenever possible.
2. Configuration in Cherry Studio:
Open Cherry Studio and navigate to Settings - MCP Servers.
Click Add Server, then select Import from JSON.
Paste the following JSON content:
{ "name": "ImgMCP", "type": "streamableHttp", "description": "Connecting Models to Your Creativity.", "provider": "ImgMCP", "providerUrl": "https://imgmcp.com", "logoUrl": "https://cdn.imgmcp.com/imgmcp-favicon.png", "baseUrl": "https://api.imgmcp.com/mcp"}
After importing, click on the newly created "ImgMCP" service in the list.
In the Request Headers section, add your API key:
Authorization=Bearer <Your API Key>
(Please be sure to replace <Your API Key> with your actual key)
Click Save.
3. Verify Connection:
Once configured, you need to verify that the MCP Host can successfully communicate with the ImgMCP platform using the new configuration.
Basic Check: In your MCP Host (e.g., Cherry Studio), check the list of MCP tools. If tools from ImgMCP are displayed correctly, the basic connection is established.
Functional Verification:
Select an appropriate LLM in your Host and enable the ImgMCP toolset.
Initiate a simple interaction. For example, you could ask: "Query historical tasks," or instruct it to perform an action that requires fetching information from the ImgMCP backend (like querying the list of available models).
If the LLM responds correctly and returns information from ImgMCP (such as the history list or model list), it indicates that your configuration is successful, and you can start using it normally.
Next Steps:
You have successfully configured your MCP Host. You can now use it to explore the various AI model interaction capabilities provided by ImgMCP. Please refer to the subsequent sections for more detailed usage examples.
The ImgMCP platform, via the MCP protocol, allows you to access and utilize various AI models. These models have different strengths, and understanding their characteristics helps you choose the right tool for specific tasks.
Main Model Types and Characteristics:
Focus on Understanding & General Purpose:
Certain models excel at interpreting complex natural language descriptions and context. For instance, models like GPT-Image-1 can handle image generation tasks involving detailed instructions or requiring contextual understanding quite well, making them suitable as general-purpose image creation tools.
Focus on Artistic Expression:
Other models stand out in rendering artistic styles and achieving unique visual effects. Midjourney, for example, is known for its strong capabilities in generating images with a distinct artistic feel or mimicking specific genres (like oil painting or printmaking), making it well-suited for artistic exploration.
Focus on Professionalism & Stability:
Another category of models is designed more towards providing stable and consistent output in specific application scenarios. Models like Flux Kontext, for instance, might perform reliably in generating product renderings or commercial advertising materials, helping to ensure professionalism and consistency in results. They are suitable for commercial or design workflows where predictable outcomes are important.
Basic Image Processing Capabilities:
In addition to the primary image generation models, we also integrate some basic image processing functions, such as:
Upscaling: To increase the resolution of existing images.
Background Removal: To separate the main subject of an image from its background.
These fundamental capabilities can supplement the core generation models, addressing common needs within the creative process.
Simplified Invocation via MCP:
A benefit of using ImgMCP is that you typically don't need to concern yourself directly with the technical details or API interfaces of each model. Through an MCP-compatible client (like Cherry Studio), you can express your requirements in natural language to a configured LLM (Large Language Model). The LLM parses this request and, via the MCP protocol, invokes the appropriate model(s) on the ImgMCP platform to perform the action. This process abstracts away the underlying complexity, allowing you to concentrate more on your creative intent.
In ImgMCP, "style" is an important concept that helps you quickly define or adjust the visual characteristics of the generated output.
What are Styles?
In the context of ImgMCP, a "style" can be understood as a series of predefined prompt fragments or parameter combinations designed to influence the visual appearance of an image. They often encapsulate common artistic genres (e.g., "watercolor," "cyberpunk"), medium textures (e.g., "claymation," "embroidery"), or lighting moods (e.g., "cinematic lighting," "golden hour").
Value of Using Styles:
Increased Efficiency: For common visual effects, selecting a preset style is usually much faster than trying to craft and refine descriptive prompts from scratch yourself.
Maintaining Consistency: When creating a series of works that require a unified visual style, using the same style settings provides convenience.
Providing Reference: Browsing the style library we provide can sometimes offer you new inspiration or direction.
Style Library:
We curate and provide a range of tested style presets covering various aesthetic directions. You can select these styles during creation as a quick start or basis for adjustment. Of course, you can (and should) still modify and combine styles based on your specific needs, or define your prompts entirely from scratch.
