Vast AI

Vast.ai lets users rent cloud GPUs for AI, machine learning, deep learning, and rendering workloads.

vast.ai
Vast AI homepage screenshot

About Vast AI

Vast.ai rents cloud GPUs for AI, machine learning, deep learning, and rendering workloads. Users add credit, search available GPUs by model, VRAM, price, and availability, then launch instances through the console, CLI, Python SDK, or REST API.

It offers on-demand GPU Cloud instances, serverless model endpoints that autoscale to zero, and dedicated multi-node clusters with InfiniBand networking. Use requires paid credit, starting at $5, with per-second billing and prices set by supply and demand.

Key features

  • Rent on-demand GPU instances filtered by model, VRAM, price, availability
  • Deploy models as serverless endpoints with autoscaling to zero
  • Dedicated multi-node clusters with InfiniBand for large training
  • Provision via CLI, Python SDK, or REST API
  • Pre-configured templates for open-source text, image, video models
  • Per-second billing with supply-and-demand pricing
Best for: Developers and AI teams that need programmable GPU rentals for training, inference, fine-tuning, and rendering.
Good for
  • Lists 20,000+ GPUs across 40+ data centers with 68+ GPU types
  • Search and deploy via console, CLI, Python SDK, or REST API
  • Serverless endpoints autoscale to zero and bill only for compute time

Vast AI at a glance

Website
vast.ai
Type
Online tool
Pricing
Paid
Platforms
Web
Open source
Not stated
Language
English
Last checked
Oct 5, 2026 · Online when checked
Listed
Oct 5, 2026

Vast AI alternatives

Tools and sites you could use instead, ranked by how closely they match.

All 24 alternatives

Popular in Cloud & Web Hosting

Well-known listings from the same topic.

All 67

Vast AI FAQ

How do I get started with Vast.ai?
Add credit starting at $5, get an API key from the console, then search for GPUs and deploy.
What deployment options are available?
GPU Cloud for on-demand instances, Serverless for endpoints that autoscale to zero, and Clusters for multi-node training.
How can I provision GPUs programmatically?
Use the CLI, Python SDK, or REST API to search, filter, and launch instances.
Is something wrong or out of date?

Tell us and an editor will check it.

Own this website?

Claim the listing and suggest updates yourself. An editor checks every claim first.

Claim this listing