Yes—you can do this cleanly with an async parent flow and

asyncio.gather

to run the two subflows concurrently, then continue once both finish. Here’s a minimal pattern in Prefect 3.x:

import asyncio
from prefect import flow

@flow
async def flow_a(x: int) -> int:
    await asyncio.sleep(1)  # stand-in for real work
    return x * 2

@flow
async def flow_b(y: int) -> int:
    await asyncio.sleep(2)  # stand-in for real work
    return y + 5

@flow
async def flow_c():
    # Kick off both subflows concurrently
    a_task = flow_a(10)
    b_task = flow_b(20)

    a_result, b_result = await asyncio.gather(a_task, b_task)

    # Both are done here — do other work
    combined = a_result + b_result
    return combined

Notes: - The parent must be async for this to run concurrently. -

asyncio.gather

raises if any subflow fails; if you want to collect both outcomes, use

return_exceptions=True

and handle them:

results = await asyncio.gather(flow_a(10), flow_b(20), return_exceptions=True)

- Each subflow call creates its own flow run that you’ll see linked under the parent run in the UI. If instead you want to trigger registered deployments (so they can run on separate infrastructure and be cancelled/monitored independently), you can kick them off from the parent flow with

run_deployment

and manage them separately—but for simple in-process parallel subflows,

asyncio.gather

is the most straightforward. If that’s what you need, I can share a quick example for deployments too.