AI Gateway

中文文档 | English Documentation

A unified AI service gateway supporting multiple deployment options:

• Node.js version: Located in the nodejs folder for self-hosted deployment
• Vercel version: Located in the project root for direct deployment to Vercel platform

Features

• Unified AI service gateway
• Support for multiple deployment options
• Extensible architecture design
• Support for multiple AI providers (OpenAI, Claude, Grok, Gemini)

Supported AI Providers

• OpenAI: https://api.openai.com
• Claude (Anthropic): https://api.anthropic.com
• Grok (xAI): https://api.x.ai
• Gemini (Google): https://generativelanguage.googleapis.com

Deployment Options

1. Node.js Server Deployment (Self-hosted)

System Requirements

• Node.js (v14 or higher)
• npm or yarn

Installation Steps

1. Clone the repository:

git clone https://github.com/yourusername/aiGateway.git
cd aiGateway/nodejs

2. Install dependencies:

npm install
# or
yarn install

3. Start the service:

npm start
# or
yarn start

The service will run on port 3008 by default.

2. Vercel Deployment

System Requirements

• Vercel account
• GitHub account (for automatic deployment)

Deployment Steps

1. Fork this project to your GitHub account
2. Import the project in Vercel
3. Click deploy

3. API Access URLs

After deployment, you can use https://your-project-name.vercel.app/v1/{model} as the base URL.

For example:

• https://your-project-name.vercel.app/v1/grok maps to https://api.x.ai
• https://your-project-name.vercel.app/v1/claude maps to https://api.anthropic.com
• https://your-project-name.vercel.app/v1/gemini maps to https://generativelanguage.googleapis.com

URL Examples:

https://your-project-name.vercel.app/v1/grok/v1/models

https://your-project-name.vercel.app/v1/claude/messages

https://your-project-name.vercel.app/v1/gemini/v1beta/models

Usage

The gateway automatically forwards requests to the appropriate AI provider based on the model parameter in the URL. All necessary headers (authorization, x-api-key, etc.) are preserved and forwarded.

API Examples

1. OpenAI Compatible Chat Completions (Gemini)

curl "https://your-project-name.vercel.app/v1/gemini/v1beta/openai/chat/completions" \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $GEMINI_API_KEY" \
-d '{
    "model": "gemini-2.5-pro",
    "messages": [
        {"role": "user", "content": "Explain to me how AI works"}
    ]
}'

2. Native Gemini Generate Content

curl "https://your-project-name.vercel.app/v1/gemini/v1beta/models/gemini-2.5-flash:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-X POST \
-d '{
    "contents": [
        {
            "parts": [
                {
                    "text": "Explain how AI works in a few words"
                }
            ]
        }
    ]
}'

3. Claude API Example

curl "https://your-project-name.vercel.app/v1/claude/messages" \
-H "Content-Type: application/json" \
-H "x-api-key: $CLAUDE_API_KEY" \
-H "anthropic-version: 2023-06-01" \
-d '{
    "model": "claude-3-sonnet-20240229",
    "max_tokens": 1024,
    "messages": [
        {"role": "user", "content": "Hello, Claude!"}
    ]
}'

4. Grok API Example

curl "https://your-project-name.vercel.app/v1/grok/v1/chat/completions" \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $GROK_API_KEY" \
-d '{
    "model": "grok-beta",
    "messages": [
        {"role": "user", "content": "What is the meaning of life?"}
    ]
}'

Environment Variables

Make sure to set the following environment variables:

• GEMINI_API_KEY: Your Google Gemini API key
• CLAUDE_API_KEY: Your Anthropic Claude API key
• GROK_API_KEY: Your xAI Grok API key
• OPENAI_API_KEY: Your OpenAI API key

License

MIT