Skip to main content
Ctrl+K

Site Navigation

  • Docs
  • AI Gallery
  • Blog
  • Community
  • Slack
  • Twitter
  • GitHub

Site Navigation

  • Docs
  • AI Gallery
  • Blog
  • Community
  • Slack
  • Twitter
  • GitHub

Getting Started

  • Installation
  • Quickstart
  • Tutorial: DNN Training
  • Start a Development Cluster

Running Jobs

  • Managed Jobs
  • Cluster Job Queue
  • Auto-provisioning GPUs
  • Running on Kubernetes
    • Getting Started
    • Kubernetes Cluster Setup
      • Deployment Guides
      • Exposing Services
    • Kubernetes Troubleshooting
  • Distributed Jobs on Many Nodes

SkyServe: Model Serving

  • Serving Models
  • Serving User Guides
    • Autoscaling
    • Updating a Service
    • Authorization
    • Using Spot Instances for Serving
  • Service YAML

Cutting Cloud Costs

  • Managed Spot Jobs
  • Autostop and Autodown
  • Benchmark: Find the Best Hardware for Your Jobs
    • CLI
    • YAML Configuration
    • SkyCallback

Using Data

  • Syncing Code and Artifacts
  • Cloud Object Storage

User Guides

  • Secrets and Environment Variables
  • Using Docker Containers
  • Opening Ports
  • Cloud TPU
  • Usage Collection
  • Frequently Asked Questions

Developer Guides

  • Contributing to SkyPilot
  • Guide: Adding a New Cloud

Cloud Admin and Usage

  • Minimal Cloud Permissions
    • AWS
    • GCP
    • vSphere
    • Kubernetes
  • Cloud Authentication
  • Requesting Quota Increase

References

  • Task YAML
  • Command Line Interface
  • Python API
  • Advanced Configurations

Serving User Guides#

  • Autoscaling
    • Fixed Replicas
    • Enabling Autoscaling
    • Scaling Delay
    • Scale-to-Zero
  • Updating a Service
    • Rolling Update
      • Example
    • Blue-Green Update
      • Example
  • Authorization
    • Setup API Keys
  • Using Spot Instances for Serving
    • Base on-demand Fallback
    • Dynamic on-demand Fallback
    • Example

previous

Serving Models

next

Autoscaling

Edit on GitHub

© Copyright 2024, SkyPilot Team.