This is a fork of the Chinese version: https://github.com/ZyphrZero/z.ai2api_python
Translated to English with enhanced admin dashboard and monitoring system.

Z.AI2API is a FastAPI-based OpenAI-compatible API server that provides access to multiple AI providers including Z.AI, K2Think, and LongCat. This English version includes a fully functional admin dashboard with real-time monitoring, request logging, and token management.

• OpenAI-compatible API - Drop-in replacement for OpenAI API
• Multi-provider support - Z.AI, K2Think, LongCat
• Token pool management - Automatic failover and load balancing
• Streaming support - Real-time response streaming
• Function calling - Tool/function call support
• Anonymous mode - Automatic temporary token acquisition

• Real-time statistics - Live request counts, success rates, uptime
• Request logging - Comprehensive logging of all API requests
• Token management - Add, validate, enable/disable tokens
• Health monitoring - Token pool health checks and status
• Configuration management - Web-based settings management
• Live monitoring - Real-time logs and provider status

• Python 3.8+
• pip or uv package manager

1. Clone the repository

   git clone <your-repo-url>
   cd z.ai2api_python

2. Install dependencies

   pip install -r requirements.txt

3. Configure environment

   cp .env.example .env
   # Edit .env with your settings

4. Start the server

   python main.py

5. Access admin dashboard

   • URL: http://localhost:8080/admin
   • Default password: admin123

• Live Statistics: Real-time request counts, success rates, service uptime
• Token Pool Health: Visual status of available authentication tokens
• Recent Requests: Live feed of API requests with success/failure status

• Add Tokens: Single or bulk token addition with validation
• Health Checks: Automatic token validation and health monitoring
• Usage Statistics: Per-token success rates and request counts
• Provider Management: Support for multiple AI providers

• Service Settings: Port, debug mode, anonymous mode
• Authentication: API keys and token management
• Advanced Options: Failure thresholds, recovery timeouts
• Hot Reload: Configuration changes without restart

• Provider Status: Real-time status of all AI providers
• Live Logs: Streaming log viewer with filtering
• System Health: Service metrics and performance monitoring

# Server Configuration
SERVICE_NAME=z-ai2api-server
LISTEN_PORT=8080
DEBUG_LOGGING=true

# Authentication
AUTH_TOKEN=sk-your-api-key
SKIP_AUTH_TOKEN=false
ANONYMOUS_MODE=true

# Features
TOOL_SUPPORT=true
SCAN_LIMIT=200000

# Token Pool
TOKEN_FAILURE_THRESHOLD=3
TOKEN_RECOVERY_TIMEOUT=1800

# Providers
DEFAULT_PROVIDER=zai
LONGCAT_TOKEN=your-longcat-token

• Default Password: admin123
• Change Password: Set ADMIN_PASSWORD in .env
• Session Timeout: 24 hours

curl -X POST http://localhost:8080/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer your-api-key" \
  -d '{
    "model": "gpt-4o",
    "messages": [
      {"role": "user", "content": "Hello, world!"}
    ],
    "stream": false
  }'

• Z.AI: GLM-4.5, GLM-4.5-Thinking, GLM-4.5-Search, GLM-4.6, etc.
• K2Think: MBZUAI-IFM/K2-Think
• LongCat: LongCat-Flash, LongCat, LongCat-Search

curl -X POST http://localhost:8080/v1/chat/completions \
  -H "Authorization: Bearer your-api-key" \
  -d '{"model": "gpt-4o", "messages": [...], "stream": true}'

The system uses SQLite for data persistence:

• tokens.db - Token management and request logs
• Automatic cleanup - Old logs removed after 30 days
• Backup recommended - Regular database backups suggested

-- Request logs table
CREATE TABLE request_logs (
    id INTEGER PRIMARY KEY AUTOINCREMENT,
    timestamp DATETIME DEFAULT CURRENT_TIMESTAMP,
    provider TEXT NOT NULL,
    model TEXT NOT NULL,
    success BOOLEAN NOT NULL,
    duration REAL,
    input_tokens INTEGER DEFAULT 0,
    output_tokens INTEGER DEFAULT 0,
    total_tokens INTEGER DEFAULT 0,
    error_message TEXT
);

-- Tokens table (managed by TokenDAO)
CREATE TABLE tokens (
    id INTEGER PRIMARY KEY AUTOINCREMENT,
    provider TEXT NOT NULL,
    token TEXT NOT NULL UNIQUE,
    token_type TEXT DEFAULT 'unknown',
    is_enabled BOOLEAN DEFAULT 1,
    created_at DATETIME DEFAULT CURRENT_TIMESTAMP
);

• All API requests are automatically logged
• Success/failure tracking with error messages
• Performance metrics including response times
• Token usage statistics per provider

• Token validation - Automatic health checks
• Provider status - Real-time availability monitoring
• System metrics - Memory, CPU, uptime tracking
• Error tracking - Comprehensive error logging

• Automatic cleanup - Old logs removed after 30 days
• Log rotation - Prevents disk space issues
• Real-time viewing - Live log streaming in dashboard

z.ai2api_python/
├── app/
│   ├── admin/          # Admin dashboard
│   ├── core/           # Core API functionality
│   ├── models/         # Database models
│   ├── providers/      # AI provider implementations
│   ├── services/       # Data access layer
│   ├── templates/      # HTML templates
│   └── utils/          # Utility functions
├── logs/               # Application logs
├── main.py            # Application entry point
└── requirements.txt   # Dependencies

1. Create provider class in app/providers/
2. Implement required methods
3. Register in provider_factory.py
4. Add model mappings
5. Update admin dashboard

• Templates: Modify files in app/templates/
• Styling: Update CSS in app/static/css/
• Update intervals: Adjust HTMX trigger timings
• Statistics: Modify API endpoints in app/admin/api.py

• All UI elements translated from Chinese to English
• Admin dashboard fully localized
• Log messages and system notifications translated
• Documentation and comments updated

• Real-time statistics - Live updating dashboard with request counts, success rates
• Request logging system - Comprehensive logging of all API requests to database
• Token pool monitoring - Visual health status of authentication tokens
• Recent request logs - Live feed showing success/failure of recent API calls
• Improved UX - Clean, professional interface with color-coded status indicators

• Live updates - Dashboard refreshes automatically without page reload
• Request tracking - All API calls logged with success/failure status
• Performance monitoring - Service uptime and response time tracking
• Error handling - Comprehensive error logging and display
• Database integration - SQLite-based persistence for logs and statistics

• HTMX integration - Seamless live updates without JavaScript complexity
• Database optimization - Indexed queries and automatic cleanup
• Error resilience - Graceful handling of database and logging failures
• Performance - Minimal impact on API response times

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

This project is licensed under the MIT License - see the LICENSE file for details.

• Original Project: ZyphrZero/z.ai2api_python
• Z.AI Platform: For providing the AI services
• FastAPI: For the excellent web framework
• HTMX: For seamless frontend interactions

Note: This is an English translation and enhancement of the original Chinese project. All core functionality remains the same with added monitoring and dashboard capabilities.