Welcome to SkyPilot!

Star Watch Fork

Run LLMs and AI on Any Cloud

SkyPilot is a framework for running LLMs, AI, and batch jobs on any cloud, offering maximum cost savings, highest GPU availability, and managed execution.

SkyPilot abstracts away cloud infra burdens:

• Launch jobs & clusters on any cloud

• Easy scale-out: queue and run many jobs, automatically managed

• Easy access to object stores (S3, GCS, R2)

SkyPilot maximizes GPU availability for your jobs:

• Provision in all zones/regions/clouds you have access to (the Sky), with automatic failover

SkyPilot cuts your cloud costs:

• Managed Spot: 3-6x cost savings using spot VMs, with auto-recovery from preemptions

• Optimizer: 2x cost savings by auto-picking the cheapest VM/zone/region/cloud

• Autostop: hands-free cleanup of idle clusters

SkyPilot supports your existing GPU, TPU, and CPU workloads, with no code changes.

Current supported providers (AWS, GCP, Azure, OCI, Lambda Cloud, RunPod, Fluidstack, Cudo, IBM, Samsung, Cloudflare, VMware vSphere, any Kubernetes cluster):

More Information

Tutorials: SkyPilot Tutorials

Runnable examples:

• LLMs on SkyPilot

  • Llama 3

  • Qwen

  • Databricks DBRX

  • Gemma

  • Mixtral 8x7B; Mistral 7B (from official Mistral team)

  • Code Llama

  • vLLM: Serving LLM 24x Faster On the Cloud (from official vLLM team)

  • SGLang: Fast and Expressive LLM Serving On the Cloud (from official SGLang team)

  • Vicuna chatbots: Training & Serving (from official Vicuna team)

  • Train your own Vicuna on Llama-2

  • Self-Hosted Llama-2 Chatbot

  • Ollama: Quantized LLMs on CPUs

  • LoRAX

  • QLoRA

  • LLaMA-LoRA-Tuner

  • Tabby: Self-hosted AI coding assistant

  • LocalGPT

  • Falcon

  • Add yours here & see more in llm/!

• Framework examples: PyTorch DDP, DeepSpeed, JAX/Flax on TPU, Stable Diffusion, Detectron2, Distributed TensorFlow, NeMo, programmatic grid search, Docker, Cog, Unsloth, Ollama and many more.

Follow updates:

• Twitter

• Slack

• SkyPilot Blog (Introductory blog post)

Read the research:

• SkyPilot paper and talk (NSDI 2023)

• Sky Computing whitepaper

• Sky Computing vision paper (HotOS 2021)

Contents

Getting Started

• Installation
• Quickstart
• Tutorial: DNN Training
• GPU-backed Jupyter Notebooks

Running Jobs

• Managed Spot Jobs
• Job Queue
• Auto-provisioning GPUs
• Running on Kubernetes
• More User Guides

SkyServe: Model Serving

• Serving Models
• Serving User Guides
• Service YAML

Cutting Cloud Costs

• Managed Spot Jobs
• Autostop and Autodown
• Benchmark: Find the Best Hardware for Your Jobs

Using Data

• Syncing Code and Artifacts
• Cloud Object Storage

User Guides

• 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
• Cloud Authentication
• Requesting Quota Increase

References

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