RunPod | Python Library

Welcome to the official Python library for RunPod API & SDK.

Table of Contents

• Table of Contents
• 💻 | Installation
• ⚡ | Serverless Worker (SDK)
  • Quick Start
  • Local Test Worker
• 📚 | API Language Library (GraphQL Wrapper)
  • Endpoints
  • GPU Cloud (Pods)
• 📁 | Directory
• 🤝 | Community and Contributing

💻 | Installation

# Install the latest release version
pip install runpod
# or# Install the latest development version (main branch)
pip install git+https://github.com/runpod/runpod-python.git

Python 3.8 or higher is required to use the latest version of this package.

⚡ | Serverless Worker (SDK)

This python package can also be used to create a serverless worker that can be deployed to RunPod as a custom endpoint API.

Quick Start

Create a python script in your project that contains your model definition and the RunPod worker start code. Run this python code as your default container start command:

# my_worker.pyimportrunpoddefis_even(job):
job_input=job["input"]
the_number=job_input["number"]
ifnotisinstance(the_number, int):
return {"error": "Silly human, you need to pass an integer."}
ifthe_number%2==0:
returnTruereturnFalserunpod.serverless.start({"handler": is_even})

Make sure that this file is ran when your container starts. This can be accomplished by calling it in the docker command when you set up a template at runpod.io/console/serverless/user/templates or by setting it as the default command in your Dockerfile.

See our blog post for creating a basic Serverless API, or view the details docs for more information.

Local Test Worker

You can also test your worker locally before deploying it to RunPod. This is useful for debugging and testing.

python my_worker.py --rp_serve_api

📚 | API Language Library (GraphQL Wrapper)

When interacting with the RunPod API you can use this library to make requests to the API.

importrunpodrunpod.api_key="your_runpod_api_key_found_under_settings"

Endpoints

You can interact with RunPod endpoints via a run or run_sync method.

endpoint=runpod.Endpoint("ENDPOINT_ID")
run_request=endpoint.run(
{"your_model_input_key": "your_model_input_value"}
)
# Check the status of the endpoint run requestprint(run_request.status())
# Get the output of the endpoint run request, blocking until the endpoint run is complete.print(run_request.output())

endpoint=runpod.Endpoint("ENDPOINT_ID")
run_request=endpoint.run_sync(
{"your_model_input_key": "your_model_input_value"}
)
# Returns the job results if completed within 90 seconds, otherwise, returns the job status.print(run_request )

GPU Cloud (Pods)

importrunpodrunpod.api_key="your_runpod_api_key_found_under_settings"# Get all my podspods=runpod.get_pods()
# Get a specific podpod=runpod.get_pod(pod.id)
# Create a podpod=runpod.create_pod("test", "runpod/stack", "NVIDIA GeForce RTX 3070")
# Stop the podrunpod.stop_pod(pod.id)
# Resume the podrunpod.resume_pod(pod.id)
# Terminate the podrunpod.terminate_pod(pod.id)

📁 | Directory

.
├── docs # Documentation
├── examples # Examples
├── runpod # Package source code
│ ├── api_wrapper # Language library - API (GraphQL)
│ ├── cli # Command Line Interface Functions
│ ├── endpoint # Language library - Endpoints
│ └── serverless # SDK - Serverless Worker
└── tests # Package tests

🤝 | Community and Contributing

We welcome both pull requests and issues on GitHub. Bug fixes and new features are encouraged, but please read our contributing guide first.