LangChain AstarCloud

A LangChain integration for AstarCloud chat models, providing a drop-in replacement for other chat model providers like OpenAI and Groq.

Installation

pip install langchain-astarcloud

Quick Start

from langchain_astarcloud import ChatAstarCloud

# Initialize the model
llm = ChatAstarCloud(
    model="gpt-4.1",
    api_key="your-api-key"  # or set ASTARCLOUD_API_KEY env var
)

# Use it like any other LangChain chat model
response = llm.invoke("Hello, how are you?")
print(response.content)

Features

• Drop-in replacement: Works seamlessly with existing LangChain and LangGraph applications
• Async support: Full async/await support for better performance
• Streaming: Real-time response streaming
• Tool calling: Support for function/tool calling on compatible models
• Type hints: Full type safety with modern Python typing

Supported Models

The following models are available:

• gpt-4.1 - Latest GPT-4 model with tool calling support
• gpt-4.1-mini - Faster, more cost-effective variant with tool calling
• gpt-4.1-nano - Ultra-fast model for simple tasks with tool calling
• astar-gpt-4.1 - AstarCloud's optimized variant with tool calling

Usage Examples

Basic Chat

from langchain_astarcloud import ChatAstarCloud

llm = ChatAstarCloud(model="gpt-4.1", api_key="sk-...")

response = llm.invoke("What's the capital of Norway?")
print(response.content)

With LangChain Expression Language (LCEL)

from langchain_core.prompts import ChatPromptTemplate
from langchain_astarcloud import ChatAstarCloud

prompt = ChatPromptTemplate.from_messages([
    ("system", "You are a helpful assistant."),
    ("human", "{question}")
])

llm = ChatAstarCloud(model="gpt-4.1")

chain = prompt | llm

response = chain.invoke({"question": "Explain quantum computing"})

Streaming Responses

from langchain_astarcloud import ChatAstarCloud

llm = ChatAstarCloud(model="gpt-4.1")

for chunk in llm.stream("Write a short story"):
    print(chunk.content, end="", flush=True)

Async Usage

import asyncio
from langchain_astarcloud import ChatAstarCloud

async def main():
    llm = ChatAstarCloud(model="gpt-4.1")
    
    response = await llm.ainvoke("Hello!")
    print(response.content)
    
    # Async streaming
    async for chunk in llm.astream("Tell me a joke"):
        print(chunk.content, end="", flush=True)

asyncio.run(main())

Tool Calling

from langchain_core.tools import tool
from langchain_astarcloud import ChatAstarCloud

@tool
def get_weather(city: str) -> str:
    """Get the weather for a city."""
    # Your weather API logic here
    return f"The weather in {city} is sunny and 22°C"

llm = ChatAstarCloud(model="gpt-4.1").bind_tools([get_weather])

response = llm.invoke("What's the weather in Oslo?")
print(response.content)
print(response.tool_calls)

With LangGraph

from langgraph.prebuilt import ToolNode
from langchain_astarcloud import ChatAstarCloud

llm = ChatAstarCloud(model="gpt-4.1").bind_tools([get_weather])

# LangGraph treats it like any other chat model
tool_node = ToolNode([get_weather])
result = llm.invoke("What's the weather in Bergen?")

Configuration

Environment Variables

Set your API key as an environment variable:

export ASTARCLOUD_API_KEY="your-api-key"

Constructor Parameters

llm = ChatAstarCloud(
    model="gpt-4.1",           # Required: Model name
    api_key="your-key",        # Optional: API key (uses env var if not provided)
    api_base="https://api.astarcloud.no",  # Optional: API base URL
    timeout=30.0,              # Optional: Request timeout in seconds
    temperature=0.7,           # Optional: Sampling temperature
    max_tokens=1000,           # Optional: Maximum tokens to generate
    # ... other model parameters
)

Error Handling

from langchain_astarcloud import ChatAstarCloud

llm = ChatAstarCloud(model="gpt-4.1")

try:
    # This will raise an error for unsupported models
    llm_with_tools = llm.bind_tools([some_tool])
    if llm.model not in {"gpt-4.1", "gpt-4.1-mini", "gpt-4.1-nano", "astar-gpt-4.1"}:
        print("Tools not supported for this model")
except ValueError as e:
    print(f"Error: {e}")

Development

To contribute to this project:

# Clone the repository
git clone https://github.com/ASTAR-CLOUD/langchain_astarcloud.git
cd langchain_astarcloud

# Install in development mode
pip install -e ".[dev]"

# Run tests
pytest

# Format code
black .
isort .

# Type checking
mypy langchain_astarcloud

License

This project is licensed under the MIT License.

Support

For support or questions:

• Open an issue on GitHub
• Contact AstarCloud support