Introduction

Cerebrium is a cloud GPU infrastructure platform that transforms how teams deploy and scale machine learning apps. By handling the complex infrastructure – GPUs, Kubernetes, queues, monitoring, and scaling – Cerebrium lets teams focus on building apps that deliver value.

The platform achieves this through rapid cold starts (Less than 5 seconds), flexible GPU options, automated scaling (1 to 10,000+ requests), and simple deployment processes. With 99.9% uptime and continuous weekly improvements based on user feedback, Cerebrium provides the reliability needed for production ML apps.

​

Getting Started

Setting up and deploying an app on Cerebrium takes just a few steps:

​

1. Install the CLI

pip install cerebrium
cerebrium login  # Redirects to your browser for authentication

​

2. Initialize a Project

cerebrium init my-first-app
cd my-first-app

This creates a basic project with main.py for app code and cerebrium.toml for configuration. This is was the main.py file contains:

def run(param_1: str, param_2: str, run_id):  # run_id is optional, injected by Cerebrium at runtime
    my_results = {"1": param_1, "2": param_2}
    my_status_code = 200

    return {"my_result": my_results, "status_code": my_status_code}

This will turn the function into a callable endpoint that accepts json parameters (param_1 and param_2) and can scale to 1000s of requests.

​

3. Deploy an App

cerebrium deploy

The app now builds and deploys, typically within a few seconds. Check out a detailed description of the deployment process for more information. Once deployed, an app becomes callable through an endpoint https://api.cortex.cerebrium.ai/v4/{project-id}/{app-name}/{function-name}.

You can also call it using:

cerebrium run my-first-app:run -d '{"param_1": "Hello", "param_2": "World"}'

​

How It Works

Cerebrium uses containerization to ensure consistent environments and reliable scaling for apps. When code is deployed, Cerebrium packages it with all necessary dependencies into a container image. This image serves as a blueprint for creating instances that handle incoming requests. The system automatically manages scaling, creating new instances when traffic increases and removing them during quiet periods.

For a detailed explanation of how Cerebrium builds and manages container images, see our Defining Container Images Guide.

Join our Discord community for support and updates, or reach out to support@cerebrium.ai with any questions.