Skip to content

[!NOTE] Active development of this project has moved within PrefectHQ/prefect. The code can be found here and documentation here. Please open issues and PRs against PrefectHQ/prefect instead of this repository.




prefect-ray contains Prefect integrations with the Ray execution framework, a flexible distributed computing framework for Python.

Provides a RayTaskRunner that enables Prefect flows to run tasks execute tasks in parallel using Ray.

Getting Started

Python setup

Requires an installation of Python 3.8 or newer.

We recommend using a Python virtual environment manager such as pipenv, conda, or virtualenv.

These tasks are designed to work with Prefect 2.0. For more information about how to use Prefect, please refer to the Prefect documentation.


Install prefect-ray with pip:

pip install prefect-ray

Users running Apple Silicon (such as M1 macs) will need to install ray in a miniforge conda environment. Click here for more details.

Running tasks on Ray

The RayTaskRunner is a Prefect task runner that submits tasks to Ray for parallel execution.

By default, a temporary Ray instance is created for the duration of the flow run.

For example, this flow counts to 3 in parallel.

import time

from prefect import flow, task
from prefect_ray import RayTaskRunner

def shout(number):

def count_to(highest_number):
    for number in range(highest_number):

if __name__ == "__main__":

# outputs

If you already have a Ray instance running, you can provide the connection URL via an address argument.

To configure your flow to use the RayTaskRunner:

  1. Make sure the prefect-ray collection is installed as described earlier: pip install prefect-ray.
  2. In your flow code, import RayTaskRunner from prefect_ray.task_runners.
  3. Assign it as the task runner when the flow is defined using the task_runner=RayTaskRunner argument.

For example, this flow uses the RayTaskRunner with a local, temporary Ray instance created by Prefect at flow run time.

from prefect import flow
from prefect_ray.task_runners import RayTaskRunner

def my_flow():

This flow uses the RayTaskRunner configured to access an existing Ray instance at ray://

from prefect import flow
from prefect_ray.task_runners import RayTaskRunner

def my_flow():

RayTaskRunner accepts the following optional parameters:

Parameter Description
address Address of a currently running Ray instance, starting with the ray:// URI.
init_kwargs Additional kwargs to use when calling ray.init.

Note that Ray Client uses the ray:// URI to indicate the address of a Ray instance. If you don't provide the address of a Ray instance, Prefect creates a temporary instance automatically.

Ray environment limitations

Ray support for non-x86/64 architectures such as ARM/M1 processors with installation from pip alone and will be skipped during installation of Prefect. It is possible to manually install the blocking component with conda. See the Ray documentation for instructions.

See the Ray installation documentation for further compatibility information.

Running tasks on a Ray remote cluster

When using the RayTaskRunner with a remote Ray cluster, you may run into issues that are not seen when using a local Ray instance. To resolve these issues, we recommend taking the following steps when working with a remote Ray cluster:

  1. By default, Prefect will not persist any data to the filesystem of the remote ray worker. However, if you want to take advantage of Prefect's caching ability, you will need to configure a remote result storage to persist results across task runs.

We recommend using the Prefect UI to configure a storage block to use for remote results storage.

Here's an example of a flow that uses caching and remote result storage:

from typing import List

from prefect import flow, get_run_logger, task
from prefect.filesystems import S3
from prefect.tasks import task_input_hash
from prefect_ray.task_runners import RayTaskRunner

# The result of this task will be cached in the configured result storage
def say_hello(name: str) -> None:
    logger = get_run_logger()
    # This log statement will print only on the first run. Subsequent runs will be cached."hello {name}!")
    return name

    # Using an S3 block that has already been created via the Prefect UI
def greetings(names: List[str]) -> None:
    for name in names:

if __name__ == "__main__":
    greetings(["arthur", "trillian", "ford", "marvin"])

  1. If you get an error stating that the module 'prefect' cannot be found, ensure prefect is installed on the remote cluster, with:

    pip install prefect

  2. If you get an error with a message similar to "File system created with scheme 's3' could not be created", ensure the required Python modules are installed on both local and remote machines. The required prerequisite modules can be found in the Prefect documentation. For example, if using S3 for the remote storage:

    pip install s3fs

  3. If you are seeing timeout or other connection errors, double check the address provided to the RayTaskRunner. The address should look similar to: address='ray://<head_node_ip_address>:10001':


Specifying remote options

The remote_options context can be used to control the task’s remote options.

For example, we can set the number of CPUs and GPUs to use for the process task:

from prefect import flow, task
from prefect_ray.task_runners import RayTaskRunner
from prefect_ray.context import remote_options

def process(x):
    return x + 1

def my_flow():
    # equivalent to setting @ray.remote(num_cpus=4, num_gpus=2)
    with remote_options(num_cpus=4, num_gpus=2):


If you encounter and bugs while using prefect-ray, feel free to open an issue in the prefect-ray repository.

If you have any questions or issues while using prefect-ray, you can find help in the Prefect Slack community.

Feel free to star or watch prefect-ray for updates, too!



If you'd like to help contribute to fix an issue or add a feature to prefect-ray, please propose changes through a pull request from a fork of the repository.

  1. Fork the repository
  2. Clone the forked repository
  3. Install the repository and its dependencies:
     pip install -e ".[dev]"
  4. Make desired changes
  5. Add tests
  6. Insert an entry to
  7. Install pre-commit to perform quality checks prior to commit: pre-commit install
  8. git commit, git push, and create a pull request