langchain-amazon-nova

Official LangChain integration for Amazon Nova models.

Amazon Nova is a family of state-of-the-art foundation models from Amazon that includes text, multimodal, and image generation capabilities. This integration provides access to Nova models through LangChain's standardized chat model interface.

Installation

pip install -U langchain-amazon-nova

Or using uv:

uv add langchain-amazon-nova

Quick Start

from langchain_amazon_nova import ChatAmazonNova

# Initialize the model
model = ChatAmazonNova(
    model="nova-pro-v1",
    temperature=0.7,
)

# Use it like any LangChain chat model
response = model.invoke("What is the capital of France?")
print(response.content)

Environment Setup

Set your Nova API credentials:

export NOVA_API_KEY="your-api-key"
export NOVA_BASE_URL="https://api.nova.amazon.com/v1"

Documentation

For detailed API documentation and additional parameters, see: https://nova.amazon.com/dev/documentation

Supported Models

Amazon Nova offers several model variants:

• nova-micro-v1: Fast, efficient text model (128K context)
• nova-lite-v1: Lightweight multimodal model (300K context)
• nova-pro-v1: Balanced multimodal model (300K context)
• nova-premier-v1: Most capable multimodal model (300K context)

Features

• Text Generation: All models support text completion
• Streaming: Token-by-token streaming responses
• Async Support: Native async/await support
• Tool Calling: Function calling capabilities (most models)
• Multimodal Input: Image and video understanding (lite, pro, premier)

Example Usage

Basic Chat

from langchain_amazon_nova import ChatAmazonNova

model = ChatAmazonNova(model="nova-pro-v1")
messages = [
    ("system", "You are a helpful assistant."),
    ("human", "Explain quantum computing in simple terms."),
]
response = model.invoke(messages)
print(response.content)

Streaming

for chunk in model.stream(messages):
    print(chunk.content, end="", flush=True)

Tool Calling

from pydantic import BaseModel, Field

class GetWeather(BaseModel):
    '''Get weather for a location.'''
    location: str = Field(description="City name")

model_with_tools = model.bind_tools([GetWeather])
response = model_with_tools.invoke("What's the weather in Tokyo?")
print(response.tool_calls)

Async

import asyncio

async def main():
    response = await model.ainvoke(messages)
    print(response.content)

asyncio.run(main())

📖 Documentation

• Integration Guide - Complete usage guide
• API Reference - Detailed API documentation
• Amazon Nova Docs - Official Nova documentation

📕 Releases & Versioning

See our Releases and Versioning policies.

💁 Contributing

As an open-source project in a rapidly developing field, we are extremely open to contributions, whether it be in the form of a new feature, improved infrastructure, or better documentation.

For detailed information on how to contribute, see the Contributing Guide.