Vast.ai is a market based cloud computing platform focused on reducing the costs and friction of compute-intensive workloads, enabling anyone to easily leverage large-scale GPU liquidity.
Our software allows all compute providers large and small to easily monetize their spare capacity.
Our search engine allows users to quickly find the best options and deals for compute services according to their specific requirements.

Vast.ai is a global market for cloud computing services. Our providers range from individual hobbyists up to tier 4 datacenters with thousands of GPUs. Our software helps streamline and automate the hosting process while allowing providers full control over pricing and rental contract details. Our interface allows searching for machines based on all relevant details including GPU type, GPU ram, CPU, system RAM, connectivity, etc.

Vast.ai's mission is to align and democratize AI. Machine learning is progressing towards powerful AI systems with the potential to radically reshape our future. We believe it is imperative that this awesome power be distributed widely; that its benefits accrue to the many rather than the few; that its secrets are unlocked for the good of all humanity. Towards these ends we work to ensure that the compute powering AI is supplied by the people and for the people.

You can talk to our team on our support chat or via email: 

. We staff the support chat 24/7 which is available in the bottom right corner here and on our 

Introduction

The Vast PyWorker is a Python web server designed to run alongside a machine learning model instance, providing autoscaler compatibility.

PyWorker

A lower level organization that lives within an Endpoint. It consists of a template (with extra filters for search), a set of GPU instances (workers) created from that template, and hyperparameters.

Worker_Group

The highest level clustering of instances for the autoscaler, consisting of a named endpoint string, a collection of Worker groups, and hyperparameters.

Endpoint

The minimum number of workers you want to keep "cold" (meaning stopped and fully loaded) when your group has no load.

min_cold_workers

The maximum number of workers your router group can have.

max_workers

QuickStart

Guides

Instances Help

Billing Help

Networking

Troubleshooting

Data Movement

Overview

Creating Templates for GROBID

Creating a Custom Template

PyTorch

AI/ML Frameworks

CUDA

GPU Programming

Linux Virtual Desktop

Linux Virtual Machines

Virtual Computing

TTS with Nari Labs Dia

AI Audio Generation

Ollama + Webui

Oobabooga (LLM webui)

Huggingface TGI with LLama3

Quantized GGUF models (cloned)

vLLM (LLM inference and serving)

AI Text Generation

Image Generation

Stable Diffusion

Disco Diffusion

AI Image Generation

Video Generation

AI Video Generation

Infinity Embeddings

Text Embeddings

Blender in the Cloud

Blender Batch Rendering

3D Rendering

Mining on Bittensor

Cryptocurrency

Google Colab

Development Tools

Whisper ASR Guide

Audio-to-Text

use-cases

Teams Overview

Edit team

Teams Quickstart

Team Creation

Teams Invitations

Teams Roles

Transfer Team Ownership

team

Guide to Taxes

Datacenter Status

Payment

Verification Stages

hosting

Create Your Own Multi-Node Cluster

distributed-computing

Settings

Keys

Instance Portal

Templates

Volumes

Billing

Earning

Instances Guide

Search

Referral Program

Members

console

RTX 5 Series

Specific GPUs