Get started with jobs

Run discrete workloads in production such as batch inference, bulk embeddings generation, or model fine-tuning.

---

Anyscale Jobs allow you to submit applications developed on workspaces to a standalone Ray cluster for execution. Built for production and designed to fit into your CI/CD pipeline, jobs ensure scalable and reliable performance.

Run your first job with the following instructions.

1. Install the Anyscale CLI

pip install -U anyscale
anyscale login

2. Submit a job

Clone the example from GitHub.

git clone https://github.com/anyscale/examples.git
cd examples/01_job_hello_world

The code in main.py runs 100 tasks that each take a number and square it.

import os
import ray


@ray.remote
def f(i):
    # This print statement is running in a separate worker process.
    print(f"The value of EXAMPLE_ENV_VAR is {os.environ['EXAMPLE_ENV_VAR']}.")
    return i ** 2


# Execute 100 tasks across the cluster.
results = ray.get([f.remote(i) for i in range(100)])
print(results)

Also take a look at job.yaml. This file specifies the container image, compute resources, script entrypoint, and a few other fields.

Submit the job:

anyscale job submit -f job.yaml

3. Inspect the results

Navigate to the Anyscale Jobs page and take a look at the results.