tnsa-api 1.1.0

A powerful, OpenAI-compatible Python SDK for TNSA NGen3 Pro and Lite Models

TNSA API Python Client

A powerful, OpenAI-compatible Python SDK for TNSA NGen3 Pro and Lite Models with MCP (Machine Control Protocol) integration.

Features

• 🚀 OpenAI-Compatible API - Familiar interface for easy migration
• ⚡ Async & Sync Support - Both synchronous and asynchronous clients
• 🌊 Streaming Responses - Real-time token streaming for interactive applications
• 🔧 Comprehensive Error Handling - Robust error handling with retry logic
• 📊 Usage Tracking - Built-in token counting and cost estimation
• 💬 Conversation Management - Automatic chat history and context management
• 🔒 Secure Authentication - API key management with environment variable support
• 📝 Type Safety - Full type hints for better IDE support
• 🎯 Framework Integration - Works seamlessly with FastAPI, Django, and more
• 🤖 MCP Client - Integrated Machine Control Protocol client for tool and service integration

Installation

pip install tnsa-api

Quick Start

Basic Usage

from tnsa_api_v2 import TNSA

# Initialize the client
client = TNSA(api_key="your-api-key")

# List available models
models = client.models.list()
print("Available models:", [model.id for model in models])

# Create a chat completion
response = client.chat.completions.create(
    model="NGen3.9-Pro",
    messages=[
        {"role": "user", "content": "Hello, how are you?"}
    ]
)

print(response.choices[0].message.content)

MCP Client Usage

The TNSA SDK includes a client for the Machine Control Protocol (MCP), enabling integration with various tools and services.

import asyncio
from tnsa_api_v2 import MCPClient

async def run_mcp_example():
    async with MCPClient(
        server_url="https://mcp.example.com",
        api_key="your-mcp-api-key"  # Optional
    ) as client:
        # List available tools
        tools = await client.list_tools()
        print("Available tools:", tools)
        
        # Call a tool
        if tools:
            result = await client.call_tool(
                tool_name=tools[0]["name"],
                params={"param1": "value1"}
            )
            print("Tool result:", result)

# Run the example
asyncio.run(run_mcp_example())

For more details, see the MCP Client Documentation.

Streaming Example

# Streaming responses
stream = client.chat.completions.create(
    model="NGen3.9-Lite",
    messages=[{"role": "user", "content": "Tell me a story"}],
    stream=True
)

for chunk in stream:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="")

Async Usage

import asyncio
from tnsa_api_v2 import AsyncTNSA

async def main():
    client = AsyncTNSA(api_key="your-api-key")
    
    response = await client.chat.completions.acreate(
        model="NGen3.9-Pro",
        messages=[{"role": "user", "content": "Hello!"}]
    )
    
    print(response.choices[0].message.content)

asyncio.run(main())

Configuration

The client can be configured using environment variables, configuration files, or direct parameters:

Environment Variables

export TNSA_API_KEY="your-api-key"
export TNSA_BASE_URL="https://api.tnsaai.com"
export TNSA_TIMEOUT=30.0

Configuration File (config.yaml)

api_key: "your-api-key"
base_url: "https://api.tnsaai.com"
timeout: 30.0
max_retries: 3
default_model: "NGen3.9-Pro"

Direct Parameters

client = TNSA(
    api_key="your-api-key",
    base_url="https://api.tnsaai.com",
    timeout=30.0,
    max_retries=3
)

Available Models

• NGen3.9-Pro - High-performance model for complex tasks
• NGen3.9-Lite - Fast, efficient model for general use
• NGen3-7B-0625 - Specialized model variant
• Farmvaidya-Bot - Agricultural domain-specific model

Error Handling

from tnsa_api_v2 import TNSAError, RateLimitError, AuthenticationError

try:
    response = client.chat.completions.create(
        model="NGen3.9-Pro",
        messages=[{"role": "user", "content": "Hello!"}]
    )
except AuthenticationError:
    print("Invalid API key")
except RateLimitError:
    print("Rate limit exceeded")
except TNSAError as e:
    print(f"API error: {e}")

Documentation

• TNSA API Documentation
• MCP Client Documentation
• Examples

Support

• 📧 Email: [e-mail]
• 🐛 Issues: GitHub Issues
• 📖 Documentation: https://docs.tnsaai.com

License

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