How do you run a blocking function inside asyncio?

Short answer

Pass it to asyncio.to_thread (or loop.run_in_executor) so it runs in a separate thread without blocking the event loop, and use a ProcessPoolExecutor with run_in_executor for CPU-bound work.

Explanation

to_thread suits blocking I/O libraries such as an old SDK or reading large files. It improves responsiveness, but it does not make Python computation parallel on traditional CPython because of the GIL. to_thread uses the default executor, which has a limited number of threads, so thousands of blocking calls will queue behind it; set an appropriate pool size or look for an asynchronous alternative.

Code example

import asyncio
from concurrent.futures import ProcessPoolExecutor

async def handler(path: str):
    data = await asyncio.to_thread(read_large_file, path)  # I/O Blocking
    loop = asyncio.get_running_loop()
    with ProcessPoolExecutor() as pool:
        digest = await loop.run_in_executor(pool, heavy_hash, data)
    return digest

Common mistakes

• Calling a blocking function directly inside a coroutine, which delays all other requests.
• Using to_thread for heavy computation and expecting real parallelism.

Follow-up questions

• What is an event loop and what is the role of a coroutine in asyncio?
• What is the effect of the GIL in CPython, and what changed with the free-threaded build?

Sources

• Python documentation — asyncio.to_thread