Instagram Linkedin X-twitter Youtube
Linkedin
X-twitter
vast-ai-logo

Vast.ai

Get in Touch
Get a Free Expert Consultation
  • 2018
  • Private
  • Los Angeles, United States
  • 11–50
  • vast.ai
Non-hyperscaler
Compute AI & GPU Infrastructure

Quick Stats

Pricing Model
Pay-As-You-Go, Spot Pricing
Customer Count
Unknown
Region
North America, South America, Europe, Asia, Oceania, Africa

About

Vast.ai offers a marketplace for affordable GPU cloud computing, enabling users to rent compute resources for AI, ML, and rendering workloads. Its unique spot-market pricing and broad hardware compatibility allow users to optimize costs and select from a wide range of global providers.

Explore
Alteranate Cloud Providers
Massed-compute-logoa
Massed Compute
backblaze-logo
Backblaze
Wasabi-Logo-600x250-1 3
Wasabi
Ionos
  • Newsletter

Stay Ahead in Cloud Infrastructure

Join the community building scalable, cost-efficient cloud strategies with insights from experts

Subscribe to Newsletter
Pricing Details

Vast.ai GPU Cloud Pricing

Marketplace for on-demand, reserved, and interruptible GPU compute
On-Demand Instances
Guaranteed resources, fixed pricing
Pay-as-you-go
GPU Compute
Marketplace pricing, varies by GPU model, region, and host.
Example: NVIDIA A100 from ~$1.20/hr, RTX 3090 from ~$0.40/hr*
Storage
Set by host, typically $0.01–$0.10/GB-month
Billed continuously while instance exists
Bandwidth
Set by host, typically $0.01–$0.10/GB transferred
Both upload and download billed
Billing granularity Per-second billing, credits required upfront
No minimum contract. Delete instances to stop storage billing.
Reserved Instances
Discounted rates with pre-payment
Up to 50% off
GPU Compute
Discounted from on-demand rates, up to 50% off
Requires pre-payment and commitment
Storage & Bandwidth Same as on-demand, set by host
Billing granularity Pre-paid, discounted hourly rate
Ideal for long-term, predictable workloads.
Interruptible Instances
Lowest cost, may be paused or preempted
Best value
GPU Compute
Often 50%+ cheaper than on-demand
May be paused or terminated by host
Storage & Bandwidth Same as on-demand, set by host
Billing granularity Per-second billing, credits required upfront
Best for non-critical, batch, or flexible workloads.

*All prices are examples and vary by GPU model, host, and region. Marketplace rates fluctuate in real time. Storage and bandwidth are set by each host. Check the Vast.ai Dashboard for current live pricing. Credits must be preloaded to your account before launching instances. Storage charges continue for stopped instances until deleted.

Free Tool

Free Cost Calculator

Deep Dive on your Cloud Storage Costs

Calculate my Costs
Features
Compute On-demand GPU instances with scalable clusters and pay-as-you-go pricing.
Performance Deploy high-performance GPU workloads for AI training and fine-tuning.
Security SOC 2 and ISO 27001-aligned environments provided by participating datacenter hosts.
Deployment Prebuilt templates for frameworks like PyTorch, TensorFlow, and CUDA; API and CLI support for automation.
Key Offerings
  • On-demand GPU cloud with instant deployment
  • Customizable GPU clusters for scalable compute needs
  • API and CLI for automated infrastructure management
  • Prebuilt templates for popular AI/ML frameworks (e.g., PyTorch, TensorFlow, CUDA)
  • Transparent pay-as-you-go pricing with spot-market discounts
Ideal Use Cases / Buyers
Workloads AI training, fine-tuning, batch GPU compute, 3D rendering, scientific simulations, and general GPU-accelerated workloads.
Buyers AI/ML engineers, data scientists, developers, researchers, startups, and enterprises requiring flexible GPU compute capacity.
Integrations & Partners
Why Choose

  • ✅ Save significantly on GPU cloud costs with transparent on-demand and spot-market pricing
  • ✅ Access thousands of GPUs across global datacenters with instant deployment
  • ✅ Flexible scaling from a single GPU to multi-GPU clusters for diverse AI workloads
  • ✅ SOC 2 and ISO 27001-aligned environments provided by participating datacenter hosts
  • ✅ API and CLI support for automation, with real-time technical assistance

Newsletter

Stay Ahead in Cloud
& Data Infrastructure

Get early access to new tools, insights, and research shaping the next wave of cloud and storage innovation.
Talk to expert

Get a free deep dive with an expert​

vast-ai-logo
Datastorage Expert
Analyst Report

Please enter your email to get this report now and on your email.