Skip to main content

​
Overview

The Reducto Python SDK provides full asynchronous support through the AsyncReducto client. This allows you to make non-blocking API calls using Python’s async/await syntax.

​
Basic Async Usage

To use the async client, import AsyncReducto instead of Reducto and use await with each API call: 
import os
import asyncio
from reducto import AsyncReducto

client = AsyncReducto(
    api_key=os.environ.get("REDUCTO_API_KEY"),  # This is the default and can be omitted
    environment="eu",  # or 'production' | 'au'; defaults to "production"
)


async def main() -> None:
    response = await client.parse.run(
        input="https://pdfobject.com/pdf/sample.pdf",
    )


asyncio.run(main())
Functionality between the synchronous and asynchronous clients is otherwise identical.

​
HTTP Backends

The async client supports two HTTP backends for different performance characteristics.

​
httpx (Default)

By default, the async client uses httpx for HTTP requests. This is included with the standard SDK installation: 
import os
import asyncio
from reducto import AsyncReducto

async def main() -> None:
    client = AsyncReducto(
        api_key=os.environ.get("REDUCTO_API_KEY"),
    )
    
    response = await client.parse.run(
        input="https://pdfobject.com/pdf/sample.pdf",
    )

asyncio.run(main())

​
aiohttp (Advanced)

For improved concurrency performance, you can use aiohttp as the HTTP backend instead of httpx.
1

Install aiohttp

Install the SDK with the aiohttp extra:
pip install reductoai[aiohttp]
2

Use DefaultAioHttpClient

Enable aiohttp by instantiating the client with http_client=DefaultAioHttpClient():
import os
import asyncio
from reducto import DefaultAioHttpClient
from reducto import AsyncReducto


async def main() -> None:
    async with AsyncReducto(
        api_key=os.environ.get("REDUCTO_API_KEY"),
        http_client=DefaultAioHttpClient(),
    ) as client:
        response = await client.parse.run(
            input="https://pdfobject.com/pdf/sample.pdf",
        )


asyncio.run(main())
When using aiohttp, it’s recommended to use the client as a context manager (with async with) to ensure proper cleanup of HTTP resources.

​
Comparison: httpx vs aiohttp

httpx

Default backend
  • Included by default
  • Good for general use
  • Simpler setup
  • No context manager required

aiohttp

Performance backend
  • Requires extra installation
  • Better concurrency performance
  • Ideal for high-throughput applications
  • Requires context manager

​
Context Manager Pattern

When using the async client, you can optionally use it as a context manager to ensure proper cleanup: 
import os
import asyncio
from reducto import AsyncReducto


async def main() -> None:
    async with AsyncReducto(
        api_key=os.environ.get("REDUCTO_API_KEY"),
    ) as client:
        response = await client.parse.run(
            input="https://pdfobject.com/pdf/sample.pdf",
        )
        # Client will automatically close when exiting the context


asyncio.run(main())

​
Multiple Concurrent Requests

The async client excels at handling multiple concurrent requests: 
import os
import asyncio
from reducto import AsyncReducto


async def main() -> None:
    client = AsyncReducto(
        api_key=os.environ.get("REDUCTO_API_KEY"),
    )
    
    # Process multiple documents concurrently
    urls = [
        "https://example.com/document1.pdf",
        "https://example.com/document2.pdf",
        "https://example.com/document3.pdf",
    ]
    
    tasks = [client.parse.run(input=url) for url in urls]
    responses = await asyncio.gather(*tasks)
    
    return responses


asyncio.run(main())
Use the async client with asyncio.gather() to process multiple documents in parallel for maximum throughput.