siliconflow-flux MCP Server

Use This MCP server To

Generate high-resolution images via AI using SiliconFlow Flux model Customize image generation parameters like resolution and random seed Cache and reuse recent image generation results for faster responses Integrate image generation capabilities into AI assistants via MCP Support multi-resolution image outputs for diverse application needs

README

SiliconFlow Flux MCP 服务器

这是一个基于Model Context Protocol (MCP)的服务器，允许AI通过MCP协议使用SiliconFlow的Flux模型进行图像生成。

功能特点

允许AI助手通过MCP协议调用SiliconFlow的Flux模型生成高质量图像
支持多种图像分辨率：1024x1024, 960x1280, 768x1024, 720x1440, 720x1280
可自定义生成参数，如随机种子等
缓存最近的图像生成结果

系统要求

Node.js v18.0.0 或更高版本

您可以通过以下命令验证Node.js安装：

node --version  # 应显示v18.0.0或更高版本

安装步骤

克隆仓库：

git clone https://github.com/lioensky/siliconflow-flux-mcp-server.git

安装依赖：

npm install

构建项目：

npm run build

配置说明

API密钥配置

在使用前，您需要：

在项目根目录创建或编辑.env文件
通过https://cloud.siliconflow.cn/i/HxM1olJu 来获取14元额度密钥
添加您的SiliconFlow API密钥：

SILICONFLOW_API_KEY=您的API密钥

注意：您需要自备SiliconFlow的API密钥才能使用此服务。

MCP服务器配置

在您的MCP配置文件中添加以下内容：

{
  "mcpServers": {
    "siliconflow-flux-mcp": {
      "command": "node",
      "args": ["路径/到/siliconflow-flux-mcp-server/build/index.js"],
      "env": {
        "SILICONFLOW_API_KEY": "您的API密钥"
      }
    }
  }
}

使用方法

配置完成后，AI助手可以通过MCP协议调用此服务生成图像。服务提供了generate_image工具，接受以下参数：

prompt：图像生成提示词（建议使用英文以获得最佳效果）
resolution：图像分辨率，支持多种标准尺寸
seed（可选）：随机种子，用于生成可重复的结果

开发者工具

您可以使用MCP Inspector工具来测试服务器：

npm run inspector

许可证

请参阅项目仓库中的许可证文件。

siliconflow-flux-mcp-server FAQ

How do I install the SiliconFlow Flux MCP server?

Clone the GitHub repo, install dependencies with npm, and build the project using npm run build.

What are the system requirements for running this MCP server?

Requires Node.js version 18.0.0 or higher and a valid SiliconFlow API key.

How do I configure the API key for the server?

Create or edit a .env file in the project root and add your SiliconFlow API key as SILICONFLOW_API_KEY.

Can I customize the image resolution and generation parameters?

Yes, the server supports multiple resolutions and customizable parameters like random seeds.

Does the server cache generated images?

Yes, it caches recent image generation results to improve response times.

How do I integrate this MCP server into my MCP client?

Add the server command and environment variables to your MCP configuration file as specified in the documentation.

Is this MCP server compatible with other LLM providers?

Yes, it works with any MCP client connected to LLMs like OpenAI, Claude, and Gemini.

Where can I get a SiliconFlow API key?

You can obtain a key by registering at https://cloud.siliconflow.cn/i/HxM1olJu with a 14 yuan credit.