Skip to content
/ zep Public
forked from skypilot-org/skypilot

Trainy Zep: Run LLMs, AI, and Batch jobs on reliably on k8s. Easy interface for running on Trainy Konduktor

License

Notifications You must be signed in to change notification settings

asaiacai/zep

 
 

Repository files navigation

Trainy Logo

This repository is a fork of the original Skypilot and maintained by Trainy in order to support running jobs Trainy's our managed Kubernetes cluster platform as a service, Konduktor (Github and Documentation). You can see some our contributions to the mainline project here. If there are features in this fork you feel like make sense to contribute back to upstream, please let us know and we are happy to make a pull request. We are planning on keeping this fork the same license as the original project (Apache 2.0), as we have also greatly benefit from the open nature of the project and believe that sharing our work reduces redundant work streams for maintainers, contributors and users alike.


SkyPilot

Documentation GitHub Release Join Slack

Run AI on Any Infra — Unified, Faster, Cheaper


🔥 News 🔥

  • [Oct 2024] 🎉 SkyPilot crossed 1M+ downloads 🎉: Thank you to our community! Twitter/X
  • [Sep 2024] Point, Launch and Serve Llama 3.2 on Kubernetes or Any Cloud: example
  • [Sep 2024] Run and deploy Pixtral, the first open-source multimodal model from Mistral AI.
  • [Jun 2024] Reproduce GPT with llm.c on any cloud: guide
  • [Apr 2024] Serve Qwen-110B on your infra: example
  • [Apr 2024] Using Ollama to deploy quantized LLMs on CPUs and GPUs: example
  • [Feb 2024] Deploying and scaling Gemma with SkyServe: example
  • [Feb 2024] Serving Code Llama 70B with vLLM and SkyServe: example
  • [Dec 2023] Mixtral 8x7B, a high quality sparse mixture-of-experts model, was released by Mistral AI! Deploy via SkyPilot on any cloud: example
  • [Nov 2023] Using Axolotl to finetune Mistral 7B on the cloud (on-demand and spot): example

LLM Finetuning Cookbooks: Finetuning Llama 2 / Llama 3.1 in your own cloud environment, privately: Llama 2 example and blog; Llama 3.1 example and blog

Archived

SkyPilot is a framework for running AI and batch workloads on any infra, offering unified execution, high cost savings, and high GPU availability.

SkyPilot abstracts away infra burdens:

SkyPilot supports multiple clusters, clouds, and hardware (the Sky):

  • Bring your reserved GPUs, Kubernetes clusters, or 12+ clouds
  • Flexible provisioning of GPUs, TPUs, CPUs, with auto-retry

SkyPilot cuts your cloud costs & maximizes GPU availability:

  • Autostop: automatic cleanup of idle resources
  • Managed Spot: 3-6x cost savings using spot instances, with preemption auto-recovery
  • Optimizer: 2x cost savings by auto-picking the cheapest & most available infra

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

Install with pip:

# Choose your clouds:
pip install -U "skypilot[kubernetes,aws,gcp,azure,oci,lambda,runpod,fluidstack,paperspace,cudo,ibm,scp]"

To get the latest features and fixes, use the nightly build or install from source:

# Choose your clouds:
pip install "skypilot-nightly[kubernetes,aws,gcp,azure,oci,lambda,runpod,fluidstack,paperspace,cudo,ibm,scp]"

Current supported infra (Kubernetes; AWS, GCP, Azure, OCI, Lambda Cloud, Fluidstack, RunPod, Cudo, Paperspace, Cloudflare, Samsung, IBM, VMware vSphere):

SkyPilot

Getting Started

You can find our documentation here.

SkyPilot in 1 Minute

A SkyPilot task specifies: resource requirements, data to be synced, setup commands, and the task commands.

Once written in this unified interface (YAML or Python API), the task can be launched on any available cloud. This avoids vendor lock-in, and allows easily moving jobs to a different provider.

Paste the following into a file my_task.yaml:

resources:
  accelerators: A100:8  # 8x NVIDIA A100 GPU

num_nodes: 1  # Number of VMs to launch

# Working directory (optional) containing the project codebase.
# Its contents are synced to ~/sky_workdir/ on the cluster.
workdir: ~/torch_examples

# Commands to be run before executing the job.
# Typical use: pip install -r requirements.txt, git clone, etc.
setup: |
  pip install "torch<2.2" torchvision --index-url https://download.pytorch.org/whl/cu121

# Commands to run as a job.
# Typical use: launch the main program.
run: |
  cd mnist
  python main.py --epochs 1

Prepare the workdir by cloning:

git clone https://github.com/pytorch/examples.git ~/torch_examples

Launch with sky launch (note: access to GPU instances is needed for this example):

sky launch my_task.yaml

SkyPilot then performs the heavy-lifting for you, including:

  1. Find the lowest priced VM instance type across different clouds
  2. Provision the VM, with auto-failover if the cloud returned capacity errors
  3. Sync the local workdir to the VM
  4. Run the task's setup commands to prepare the VM for running the task
  5. Run the task's run commands

SkyPilot Demo

Refer to Quickstart to get started with SkyPilot.

More Information

To learn more, see our documentation, blog, and community integrations.

Runnable examples:

Case Studies and Integrations: Community Spotlights

Follow updates:

Read the research:

Support and Questions

We are excited to hear your feedback!

For general discussions, join us on the SkyPilot Slack.

Contributing

We welcome all contributions to the project! See CONTRIBUTING for how to get involved.

About

Trainy Zep: Run LLMs, AI, and Batch jobs on reliably on k8s. Easy interface for running on Trainy Konduktor

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 95.6%
  • Jinja 2.4%
  • Shell 1.5%
  • Other 0.5%