langflow-python-client 1.0.0

Async Python client library for interacting with Langflow API servers

Langflow Python Client

An async Python client library for interacting with Langflow API servers.

Features

Features

• 🚀 Async/await support - Built with modern Python async patterns for high performance
• 🔐 Authentication - Secure API key authentication for protected Langflow instances
• 🎛️ Dynamic Tweaks - Modify flow behavior at runtime without changing flow definitions
• 📁 File Upload - Upload files directly to flows for document processing and analysis
• 📝 Type Safety - Full type annotation support with IDE autocomplete
• 🔧 Flexible Configuration - Support for different input/output types (text, chat)
• 🐍 Python 3.10+ - Compatible with modern Python versions

Installation

pip install langflow-python-client

Quick Start

Basic Usage

import asyncio
from langflow_client import LangflowClient, FlowRequestOptions
from langflow_client.constants import InputTypes, OutputTypes

async def main():
    # Initialize the client
    client = LangflowClient(
        base_url="http://localhost:7860"
    )
    
    # Get a flow instance
    flow = client.flow("your-flow-id")
    
    # Create request options
    options = FlowRequestOptions(
        input_type=InputTypes.TEXT,
        output_type=OutputTypes.TEXT,
        session_id="my-session"
    )
    
    # Run the flow
    result = await flow.run(
        "Hello, Langflow!",
        options=options
    )
    
    print(result)

# Run the async function
asyncio.run(main())

Using Tweaks for Dynamic Configuration

Tweaks allow you to modify flow components at runtime without changing the flow definition:

async def run_with_tweaks():
    client = LangflowClient(base_url="http://localhost:7860")
    
    # Create flow with component tweaks
    flow = client.flow("flow-id").tweak(**{
        "TextInput-ABC123": {
            "input_value": "Candidate CV content here..."
        },
        "OpenAI-DEF456": {
            "temperature": 0.7,
            "max_tokens": 500
        },
        "ChatOutput-GHI789": {
            "should_store_message": False
        }
    })
    
    # Run with options
    options = FlowRequestOptions(
        input_type=InputTypes.CHAT,
        output_type=OutputTypes.CHAT
    )
    
    result = await flow.run(options=options)
    return result

Input and Output Types

The client supports different input and output types depending on your Langflow components:

Type	Description	Use Case
InputTypes.TEXT	Plain text input	Simple text processing, document analysis
InputTypes.CHAT	Chat message format	Interactive chatbots, conversational AI
OutputTypes.TEXT	Plain text output	Clean text responses
OutputTypes.CHAT	Chat message format	Interactive chatbots, conversational

License

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