# Directory Structure
```
├── .gitignore
├── Dockerfile
├── LICENSE.txt
├── package-lock.json
├── package.json
├── README.md
├── smithery.yaml
├── src
│ └── index.ts
└── tsconfig.json
```
# Files
--------------------------------------------------------------------------------
/.gitignore:
--------------------------------------------------------------------------------
```
1 | # Dependencies
2 | node_modules/
3 | npm-debug.log*
4 | yarn-debug.log*
5 | yarn-error.log*
6 |
7 | # Build
8 | build/
9 | dist/
10 | *.tsbuildinfo
11 |
12 | # Environment
13 | .env
14 | .env.local
15 | .env.*.local
16 |
17 | # IDE
18 | .idea/
19 | .vscode/
20 | *.swp
21 | *.swo
22 |
23 | # OS
24 | .DS_Store
25 | Thumbs.db
```
--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
```markdown
1 | [](https://mseep.ai/app/66julienmartin-mcp-server-qwen-max)
2 |
3 | # Qwen Max MCP Server
4 |
5 | A Model Context Protocol (MCP) server implementation for the Qwen Max language model.
6 |
7 | [](https://smithery.ai/server/@66julienmartin/mcp-server-qwen_max)
8 |
9 | <a href="https://glama.ai/mcp/servers/1v7po9oa9w"><img width="380" height="200" src="https://glama.ai/mcp/servers/1v7po9oa9w/badge" alt="Qwen Max Server MCP server" /></a>
10 |
11 | Why Node.js?
12 | This implementation uses Node.js/TypeScript as it currently provides the most stable and reliable integration
13 | with MCP servers compared to other languages like Python. The Node.js SDK for MCP offers better type safety,
14 | error handling, and compatibility with Claude Desktop.
15 |
16 | ## Prerequisites
17 |
18 | - Node.js (v18 or higher)
19 | - npm
20 | - Claude Desktop
21 | - Dashscope API key
22 |
23 | ## Installation
24 |
25 | ### Installing via Smithery
26 |
27 | To install Qwen Max MCP Server for Claude Desktop automatically via [Smithery](https://smithery.ai/server/@66julienmartin/mcp-server-qwen_max):
28 |
29 | ```bash
30 | npx -y @smithery/cli install @66julienmartin/mcp-server-qwen_max --client claude
31 | ```
32 |
33 | ### Manual Installation
34 | ```bash
35 | git clone https://github.com/66julienmartin/mcp-server-qwen-max.git
36 | cd Qwen_Max
37 | npm install
38 | ```
39 |
40 | ## Model Selection
41 | By default, this server uses the Qwen-Max model.
42 | The Qwen series offers several commercial models with different capabilities:
43 |
44 | ### Qwen-Max
45 | Provides the best inference performance, especially for complex and multi-step tasks.
46 |
47 | Context window: 32,768 tokens
48 | - Max input: 30,720 tokens
49 | - Max output: 8,192 tokens
50 | - Pricing: $0.0016/1K tokens (input), $0.0064/1K tokens (output)
51 | - Free quota: 1 million tokens
52 |
53 | Available versions:
54 |
55 | - qwen-max (Stable)
56 | - qwen-max-latest (Latest)
57 | - qwen-max-2025-01-25 (Snapshot, also known as qwen-max-0125 or Qwen2.5-Max)
58 |
59 | ### Qwen-Plus
60 | Balanced combination of performance, speed, and cost, ideal for moderately complex tasks.
61 |
62 | Context window: 131,072 tokens
63 | - Max input: 129,024 tokens
64 | - Max output: 8,192 tokens
65 | - Pricing: $0.0004/1K tokens (input), $0.0012/1K tokens (output)
66 | - Free quota: 1 million tokens
67 |
68 | Available versions:
69 |
70 | - qwen-plus (Stable)
71 | - qwen-plus-latest (Latest)
72 | - qwen-plus-2025-01-25 (Snapshot, also known as qwen-plus-0125)
73 |
74 | ### Qwen-Turbo
75 | Fast speed and low cost, suitable for simple tasks.
76 |
77 | - Context window: 1,000,000 tokens
78 | - Max input: 1,000,000 tokens
79 | - Max output: 8,192 tokens
80 | - Pricing: $0.00005/1K tokens (input), $0.0002/1K tokens (output)
81 | - Free quota: 1 million tokens
82 |
83 | Available versions:
84 |
85 | - qwen-turbo (Stable)
86 | - qwen-turbo-latest (Latest)
87 | - qwen-turbo-2024-11-01 (Snapshot, also known as qwen-turbo-1101)
88 |
89 | To modify the model, update the model name in src/index.ts:
90 |
91 | ```typescript
92 | // For Qwen-Max (default)
93 | model: "qwen-max"
94 |
95 | // For Qwen-Plus
96 | model: "qwen-plus"
97 |
98 | // For Qwen-Turbo
99 | model: "qwen-turbo"
100 | ```
101 |
102 | For more detailed information about available models, visit the Alibaba Cloud Model Documentation https://www.alibabacloud.com/help/en/model-studio/getting-started/models?spm=a3c0i.23458820.2359477120.1.446c7d3f9LT0FY.
103 |
104 | ## Project Structure
105 | ```
106 | qwen-max-mcp/
107 | ├── src/
108 | │ ├── index.ts # Main server implementation
109 | ├── build/ # Compiled files
110 | │ ├── index.js
111 | ├── LICENSE
112 | ├── README.md
113 | ├── package.json
114 | ├── package-lock.json
115 | └── tsconfig.json
116 | ```
117 | ## Configuration
118 |
119 | 1. Create a `.env` file in the project root:
120 | ```
121 | DASHSCOPE_API_KEY=your-api-key-here
122 | ```
123 |
124 | 2. Update Claude Desktop configuration:
125 | ```json
126 | {
127 | "mcpServers": {
128 | "qwen_max": {
129 | "command": "node",
130 | "args": ["/path/to/Qwen_Max/build/index.js"],
131 | "env": {
132 | "DASHSCOPE_API_KEY": "your-api-key-here"
133 | }
134 | }
135 | }
136 | }
137 | ```
138 |
139 | ## Development
140 |
141 | ```bash
142 | npm run dev # Watch mode
143 | npm run build # Build
144 | npm run start # Start server
145 | ```
146 |
147 | ## Features
148 |
149 | - Text generation with Qwen models
150 | - Configurable parameters (max_tokens, temperature)
151 | - Error handling
152 | - MCP protocol support
153 | - Claude Desktop integration
154 | - Support for all Qwen commercial models (Max, Plus, Turbo)
155 | - Extensive token context windows
156 |
157 | ## API Usage
158 |
159 | ```typescript
160 | // Example tool call
161 | {
162 | "name": "qwen_max",
163 | "arguments": {
164 | "prompt": "Your prompt here",
165 | "max_tokens": 8192,
166 | "temperature": 0.7
167 | }
168 | }
169 | ```
170 | ## The Temperature Parameter
171 |
172 | The temperature parameter controls the randomness of the model's output:
173 |
174 | Lower values (0.0-0.7): More focused and deterministic outputs
175 | Higher values (0.7-1.0): More creative and varied outputs
176 |
177 | Recommended temperature settings by task:
178 |
179 | Code generation: 0.0-0.3
180 | Technical writing: 0.3-0.5
181 | General tasks: 0.7 (default)
182 | Creative writing: 0.8-1.0
183 |
184 | ## Error Handling
185 |
186 | The server provides detailed error messages for common issues:
187 |
188 | API authentication errors
189 | Invalid parameters
190 | Rate limiting
191 | Network issues
192 | Token limit exceeded
193 | Model availability issues
194 |
195 | ## Contributing
196 | Contributions are welcome! Please feel free to submit a Pull Request.
197 |
198 | ## License
199 | MIT
```
--------------------------------------------------------------------------------
/tsconfig.json:
--------------------------------------------------------------------------------
```json
1 | {
2 | "compilerOptions": {
3 | "target": "ES2022",
4 | "module": "Node16",
5 | "moduleResolution": "Node16",
6 | "outDir": "./build",
7 | "rootDir": "./src",
8 | "strict": true,
9 | "esModuleInterop": true,
10 | "skipLibCheck": true,
11 | "forceConsistentCasingInFileNames": true
12 | },
13 | "files": ["./src/index.ts"],
14 | "include": ["src/**/*.ts"],
15 | "exclude": ["node_modules"]
16 | }
```
--------------------------------------------------------------------------------
/package.json:
--------------------------------------------------------------------------------
```json
1 | {
2 | "name": "qwen_max",
3 | "version": "1.0.0",
4 | "type": "module",
5 | "bin": {
6 | "qwen_max": "./build/index.js"
7 | },
8 | "scripts": {
9 | "build": "tsc",
10 | "start": "node build/index.js",
11 | "dev": "tsc --watch"
12 | },
13 | "dependencies": {
14 | "@modelcontextprotocol/sdk": "0.6.0",
15 | "dotenv": "^16.4.7",
16 | "openai": "^4.80.1"
17 | },
18 | "devDependencies": {
19 | "@types/node": "^20.11.24",
20 | "typescript": "^5.3.3"
21 | }
22 | }
```
--------------------------------------------------------------------------------
/smithery.yaml:
--------------------------------------------------------------------------------
```yaml
1 | # Smithery configuration file: https://smithery.ai/docs/config#smitheryyaml
2 |
3 | startCommand:
4 | type: stdio
5 | configSchema:
6 | # JSON Schema defining the configuration options for the MCP.
7 | type: object
8 | required:
9 | - dashscopeApiKey
10 | properties:
11 | dashscopeApiKey:
12 | type: string
13 | description: The API key for the Dashscope server.
14 | commandFunction:
15 | # A function that produces the CLI command to start the MCP on stdio.
16 | |-
17 | config => ({ command: 'node', args: ['build/index.js'], env: { DASHSCOPE_API_KEY: config.dashscopeApiKey } })
18 |
```
--------------------------------------------------------------------------------
/Dockerfile:
--------------------------------------------------------------------------------
```dockerfile
1 | # Generated by https://smithery.ai. See: https://smithery.ai/docs/config#dockerfile
2 | # Use Node.js 18 as the base image
3 | FROM node:18-alpine AS builder
4 |
5 | # Set working directory
6 | WORKDIR /app
7 |
8 | # Copy package files
9 | COPY package.json package-lock.json ./
10 |
11 | # Install dependencies
12 | RUN npm install
13 |
14 | # Copy the rest of the application code
15 | COPY src ./src
16 | COPY tsconfig.json ./
17 |
18 | # Build the application
19 | RUN npx tsc
20 |
21 | # Create the runtime image
22 | FROM node:18-alpine
23 |
24 | WORKDIR /app
25 |
26 | # Copy built application from the builder stage
27 | COPY --from=builder /app/build ./build
28 | COPY package.json package-lock.json ./
29 |
30 | # Install production dependencies
31 | RUN npm install --production
32 |
33 | # Copy .env file (or ensure it's mounted at runtime)
34 | COPY .env ./
35 |
36 | # Expose necessary ports (if any are known; adjust as needed)
37 | EXPOSE 3000
38 |
39 | # Define the default command
40 | CMD ["node", "build/index.js"]
41 |
```
--------------------------------------------------------------------------------
/LICENSE.txt:
--------------------------------------------------------------------------------
```
1 | MIT License
2 |
3 | Copyright (c) 2025 Kamel IRZOUNI
4 |
5 | Permission is hereby granted, free of charge, to any person obtaining a copy
6 | of this software and associated documentation files (the "Software"), to deal
7 | in the Software without restriction, including without limitation the rights
8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
9 | copies of the Software, and to permit persons to whom the Software is
10 | furnished to do so, subject to the following conditions:
11 |
12 | The above copyright notice and this permission notice shall be included in all
13 | copies or substantial portions of the Software.
14 |
15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21 | SOFTWARE.
```
--------------------------------------------------------------------------------
/src/index.ts:
--------------------------------------------------------------------------------
```typescript
1 | #!/usr/bin/env node
2 | import { Server } from "@modelcontextprotocol/sdk/server/index.js";
3 | import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";
4 | import {
5 | ListToolsRequestSchema,
6 | CallToolRequestSchema,
7 | ErrorCode,
8 | McpError
9 | } from "@modelcontextprotocol/sdk/types.js";
10 | import OpenAI from "openai";
11 | import dotenv from "dotenv";
12 |
13 | dotenv.config();
14 |
15 | const QWEN_BASE_URL = "https://dashscope-intl.aliyuncs.com/compatible-mode/v1";
16 | const API_KEY = process.env.DASHSCOPE_API_KEY;
17 |
18 | if (!API_KEY) {
19 | throw new Error("DASHSCOPE_API_KEY environment variable is required");
20 | }
21 |
22 | interface QwenMaxArgs {
23 | prompt: string;
24 | max_tokens?: number;
25 | temperature?: number;
26 | }
27 |
28 | class QwenMaxServer {
29 | private server: Server;
30 | private openai: OpenAI;
31 |
32 | constructor() {
33 | this.server = new Server(
34 | { name: "qwen_max", version: "1.0.0" },
35 | { capabilities: { tools: {} } }
36 | );
37 |
38 | this.openai = new OpenAI({
39 | apiKey: API_KEY,
40 | baseURL: QWEN_BASE_URL
41 | });
42 |
43 | this.setupHandlers();
44 | this.setupErrorHandling();
45 | }
46 |
47 | private setupErrorHandling(): void {
48 | this.server.onerror = (error: Error): void => {
49 | console.error("[MCP Error]", error);
50 | };
51 |
52 | process.on("SIGINT", async (): Promise<void> => {
53 | await this.server.close();
54 | process.exit(0);
55 | });
56 | }
57 |
58 | private setupHandlers(): void {
59 | this.server.setRequestHandler(
60 | ListToolsRequestSchema,
61 | async () => ({
62 | tools: [{
63 | name: "qwen_max",
64 | description: "Generate text using Qwen Max model",
65 | inputSchema: {
66 | type: "object",
67 | properties: {
68 | prompt: {
69 | type: "string",
70 | description: "The text prompt to generate content from"
71 | },
72 | max_tokens: {
73 | type: "number",
74 | description: "Maximum number of tokens to generate",
75 | default: 8192
76 | },
77 | temperature: {
78 | type: "number",
79 | description: "Sampling temperature (0-2)",
80 | default: 0.7,
81 | minimum: 0,
82 | maximum: 2
83 | }
84 | },
85 | required: ["prompt"]
86 | }
87 | }]
88 | })
89 | );
90 |
91 | this.server.setRequestHandler(
92 | CallToolRequestSchema,
93 | async (request) => {
94 | if (request.params.name !== "qwen_max") {
95 | throw new McpError(
96 | ErrorCode.MethodNotFound,
97 | `Unknown tool: ${request.params.name}`
98 | );
99 | }
100 |
101 | const { prompt, max_tokens = 8192, temperature = 0.7 } =
102 | request.params.arguments as QwenMaxArgs;
103 |
104 | try {
105 | const completion = await this.openai.chat.completions.create({
106 | model: "qwen-max-latest",
107 | messages: [{ role: "user", content: prompt }],
108 | max_tokens,
109 | temperature
110 | });
111 |
112 | return {
113 | content: [{
114 | type: "text",
115 | text: completion.choices[0].message.content || ""
116 | }]
117 | };
118 | } catch (error: any) {
119 | console.error("Qwen API Error:", error);
120 | throw new McpError(
121 | ErrorCode.InternalError,
122 | `Qwen API error: ${error.message}`
123 | );
124 | }
125 | }
126 | );
127 | }
128 |
129 | async run(): Promise<void> {
130 | const transport = new StdioServerTransport();
131 | await this.server.connect(transport);
132 | console.error("Qwen Max MCP server running on stdio");
133 | }
134 | }
135 |
136 | const server = new QwenMaxServer();
137 | server.run().catch(console.error);
```