Vast.ai website screenshot

Vast.ai

Vast.ai is a decentralized GPU marketplace that matches buyers with 40+ datacenters and individual hosts renting out NVIDIA GPUs at market prices. It exposes a REST API, Python SDK, and vastai CLI for searching GPU offers, renting instances, attaching persistent volumes, and running serverless workloads across 68+ GPU types including H100, H200, B200, A100, and consumer RTX cards.

Vast.ai publishes 10 APIs on the APIs.io network, including Accounts API, Billing API, Instances API, and 7 more. Tagged areas include AI, Cloud, Compute, GPU, and Machine Learning.

Vast.ai’s developer surface includes authentication, documentation, developer portal, signup flow, pricing, engineering blog, support, and 14 more developer resources.

52.4/100 developing ▬ flat Agent 44/100 agent ready Full breakdown ↓
scored 2026-07-28 · rubric v0.6
AccessFreeSelf serve⚡ Free to try
11 APIs 6 Features
AICloudComputeGPUMachine LearningMarketplaceServerless

Kin Score

Kin Score Kin Score How this is scored →
scored 2026-07-28 · rubric v0.6
Composite quality — 52.4/100 · developing
Contract Quality 15.7 / 25
Developer Ergonomics 9.6 / 20
Commercial Clarity 16.3 / 20
Operational Transparency 3.4 / 13
Governance 0.0 / 12
Discoverability 7.4 / 10
Agent readiness — 44/100 · agent ready
Machine-Readable Contract 18 / 18
Agentic Access Contract 10 / 10
MCP Server 0 / 12
Machine-Readable Auth 10 / 10
Idempotency 0 / 9
Stable Error Semantics 8 / 8
Request/Response Examples 7 / 7
Rate-Limit Signaling 7 / 7
Typed Event Surface 0 / 6
Agent Skills 0 / 5
Well-Known Catalog 0 / 4
Consent & Bot Identity 0 / 3
A2A Agent Card 0 / 8
Dry-Run / Simulate Mode 0 / 4
Improve this rating by publishing the missing artifacts — every area above can be raised, and the full rubric is at apis.io/rating/. This rating is computed from github.com/api-evangelist/vast-ai: open an issue to ask a question, or submit a pull request to add artifacts. Want it done for you? Prioritized profiling — $2,500 →

APIs 11

Individual APIs this provider publishes, each with its own machine-readable definition.

Vast.ai REST API

The Vast.ai REST API is the control plane for the GPU marketplace. It supports searching offers across hosts, renting and managing instances, configuring machines and templates,...

Vast.ai Accounts API

The Accounts API from Vast.ai — 13 operation(s) for accounts.

Vast.ai Billing API

The Billing API from Vast.ai — 4 operation(s) for billing.

Vast.ai Instances API

The Instances API from Vast.ai — 14 operation(s) for instances.

Vast.ai Machines API

The Machines API from Vast.ai — 10 operation(s) for machines.

Vast.ai Network Volumes API

The Network Volumes API from Vast.ai — 4 operation(s) for network volumes.

Vast.ai Search API

The Search API from Vast.ai — 3 operation(s) for search.

Vast.ai Serverless API

The Serverless API from Vast.ai — 9 operation(s) for serverless.

Vast.ai Team API

The Team API from Vast.ai — 9 operation(s) for team.

Vast.ai Templates API

The Templates API from Vast.ai — 1 operation(s) for templates.

Vast.ai Volumes API

The Volumes API from Vast.ai — 4 operation(s) for volumes.

Scroll for all 11

Open Collections 1

Open, tool-agnostic API collections (OpenAPI-derived and Bruno).

Vast.ai API

OPEN COLLECTION

GraphQL 1

GraphQL schemas published by this provider.

Vast.ai GraphQL API

Vast.ai is a decentralized GPU marketplace for AI and ML workloads. The API covers instance rental, machine offers, container management, SSH access configuration, and billing f...

GRAPHQL

Pricing Plans 1

Published pricing tiers and plan structures.

Vast Ai Plans Pricing

1 plans

PLANS

Rate Limits 1

Documented rate limits and quota policies.

Vast Ai Rate Limits

2 limits

RATE LIMITS

FinOps 1

Cost, billing, and metering signals for API financial operations.

Features 6

Notable capabilities this provider offers.

GPU Marketplace

Market-driven matching of buyers with 40+ datacenters and individual hosts across 68+ GPU types.

On-Demand Instances

Per-second-billed GPU rentals with guaranteed uptime, with no minimum hours.

Interruptible Instances

Fault-tolerant workloads priced 50%+ cheaper than on-demand with potential preemption.

Reserved Capacity

Up to 50% discounts on 1, 3, or 6-month commitments for stable workloads.

Serverless Endpoints

Serverless GPU workers built on the pyworker agent for autoscaling inference.

Templates

Pre-configured instance templates for fast launching of common workloads.

Security Posture 3

Authentication, domain security, vulnerability disclosure, and trust-center signals.

Vast Ai Authentication

apiKey/http · 3 schemes

SECURITY

Vast Ai Domain Security

TLSv1.3 · HSTS · DMARC

SECURITY

Vast Ai Trust Center

SOC 2, ISO 27001, PCI DSS, HIPAA, GDPR

SECURITY

Agentic Access 1

Recommended x-agentic-access execution contracts for AI agents.

Vast Ai Agentic Access

89 operations · 64 acting · 2 human-in-the-loop

89 operations · 64 acting

AGENTIC

Integrations 2

Pre-built integrations with other platforms and tools.

Docker

Bring-your-own Docker image support across all rental instances.

SSH

Direct SSH access to rented instances for development and orchestration.

Resources

Get Started 3

Portal, sign-up, and the first successful call

Documentation 1

Reference material describing how the API behaves

Agent Surfaces 2

MCP servers, agent skills, and machine-readable catalogs

Build 4

SDKs, sample code, and the tooling you integrate with

Access & Security 3

Authentication, authorization, and security posture

Operate 1

Status, limits, changes, and where to get help

Commercial 3

Pricing, plans, and the legal terms of use

Company 3

The organization behind the API

Other 2

Properties that don't map to a standard resource type

Source (apis.yml)

apis.yml Raw ↑
aid: vast-ai
name: Vast.ai
description: Vast.ai is a decentralized GPU marketplace that matches buyers with 40+ datacenters and individual hosts renting
  out NVIDIA GPUs at market prices. It exposes a REST API, Python SDK, and vastai CLI for searching GPU offers, renting instances,
  attaching persistent volumes, and running serverless workloads across 68+ GPU types including H100, H200, B200, A100, and
  consumer RTX cards.
accessModel:
  pricing: free
  onboarding: self-serve
  trial: false
  try_now: true
  public: false
  label: Free · Self-serve signup
  confidence: high
  source:
  - plans
  - authentication
  generated: '2026-07-22'
  method: derived
image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/icons/vast-ai.png
url: https://raw.githubusercontent.com/api-evangelist/vast-ai/refs/heads/main/apis.yml
created: '2026-05-23'
modified: '2026-05-23'
specificationVersion: '0.19'
type: Index
access: 3rd-Party
position: Producer
tags:
- AI
- Cloud
- Compute
- GPU
- Machine Learning
- Marketplace
- Serverless
apis:
- aid: vast-ai:rest-api
  name: Vast.ai REST API
  description: The Vast.ai REST API is the control plane for the GPU marketplace. It supports searching offers across hosts,
    renting and managing instances, configuring machines and templates, attaching persistent volumes, managing serverless
    endpoints, and querying billing and usage. Authentication uses bearer tokens issued from the Keys management page.
  humanURL: https://docs.vast.ai/api/
  baseURL: https://console.vast.ai/api/v0
  tags:
  - Billing
  - GPU
  - Instances
  - Marketplace
  - REST
  - Search
  - Serverless
  - Templates
  - Volumes
  properties:
  - type: Documentation
    url: https://docs.vast.ai/api/
  - type: Authentication
    url: https://docs.vast.ai/api/authentication
  - url: graphql/vast-ai-graphql.md
    type: GraphQL
- aid: vast-ai:vast-ai-accounts-api
  name: Vast.ai Accounts API
  description: The Accounts API from Vast.ai — 13 operation(s) for accounts.
  humanURL: https://docs.vast.ai/api/
  baseURL: https://console.vast.ai/api/v0
  tags:
  - Accounts
  properties:
  - type: OpenAPI
    url: openapi/vast-ai-accounts-api-openapi.yml
- aid: vast-ai:vast-ai-billing-api
  name: Vast.ai Billing API
  description: The Billing API from Vast.ai — 4 operation(s) for billing.
  humanURL: https://docs.vast.ai/api/
  baseURL: https://console.vast.ai/api/v0
  tags:
  - Billing
  properties:
  - type: OpenAPI
    url: openapi/vast-ai-billing-api-openapi.yml
- aid: vast-ai:vast-ai-instances-api
  name: Vast.ai Instances API
  description: The Instances API from Vast.ai — 14 operation(s) for instances.
  humanURL: https://docs.vast.ai/api/
  baseURL: https://console.vast.ai/api/v0
  tags:
  - Instances
  properties:
  - type: OpenAPI
    url: openapi/vast-ai-instances-api-openapi.yml
- aid: vast-ai:vast-ai-machines-api
  name: Vast.ai Machines API
  description: The Machines API from Vast.ai — 10 operation(s) for machines.
  humanURL: https://docs.vast.ai/api/
  baseURL: https://console.vast.ai/api/v0
  tags:
  - Machines
  properties:
  - type: OpenAPI
    url: openapi/vast-ai-machines-api-openapi.yml
- aid: vast-ai:vast-ai-network-volumes-api
  name: Vast.ai Network Volumes API
  description: The Network Volumes API from Vast.ai — 4 operation(s) for network volumes.
  humanURL: https://docs.vast.ai/api/
  baseURL: https://console.vast.ai/api/v0
  tags:
  - Network Volumes
  properties:
  - type: OpenAPI
    url: openapi/vast-ai-network-volumes-api-openapi.yml
- aid: vast-ai:vast-ai-search-api
  name: Vast.ai Search API
  description: The Search API from Vast.ai — 3 operation(s) for search.
  humanURL: https://docs.vast.ai/api/
  baseURL: https://console.vast.ai/api/v0
  tags:
  - Search
  properties:
  - type: OpenAPI
    url: openapi/vast-ai-search-api-openapi.yml
- aid: vast-ai:vast-ai-serverless-api
  name: Vast.ai Serverless API
  description: The Serverless API from Vast.ai — 9 operation(s) for serverless.
  humanURL: https://docs.vast.ai/api/
  baseURL: https://console.vast.ai/api/v0
  tags:
  - Serverless
  properties:
  - type: OpenAPI
    url: openapi/vast-ai-serverless-api-openapi.yml
- aid: vast-ai:vast-ai-team-api
  name: Vast.ai Team API
  description: The Team API from Vast.ai — 9 operation(s) for team.
  humanURL: https://docs.vast.ai/api/
  baseURL: https://console.vast.ai/api/v0
  tags:
  - Team
  properties:
  - type: OpenAPI
    url: openapi/vast-ai-team-api-openapi.yml
- aid: vast-ai:vast-ai-templates-api
  name: Vast.ai Templates API
  description: The Templates API from Vast.ai — 1 operation(s) for templates.
  humanURL: https://docs.vast.ai/api/
  baseURL: https://console.vast.ai/api/v0
  tags:
  - Templates
  properties:
  - type: OpenAPI
    url: openapi/vast-ai-templates-api-openapi.yml
- aid: vast-ai:vast-ai-volumes-api
  name: Vast.ai Volumes API
  description: The Volumes API from Vast.ai — 4 operation(s) for volumes.
  humanURL: https://docs.vast.ai/api/
  baseURL: https://console.vast.ai/api/v0
  tags:
  - Volumes
  properties:
  - type: OpenAPI
    url: openapi/vast-ai-volumes-api-openapi.yml
common:
- type: AgenticAccess
  url: agentic-access/vast-ai-agentic-access.yml
- type: TrustCenter
  url: security/vast-ai-trust-center.yml
- type: DomainSecurity
  url: security/vast-ai-domain-security.yml
- type: Authentication
  url: authentication/vast-ai-authentication.yml
- type: Website
  url: https://vast.ai
- type: Developer
  url: https://docs.vast.ai
- type: Documentation
  url: https://docs.vast.ai
- type: Portal
  url: https://cloud.vast.ai
- type: Signup
  url: https://cloud.vast.ai/create
- type: Login
  url: https://cloud.vast.ai/login
- type: Pricing
  url: https://vast.ai/pricing
- type: Blog
  url: https://vast.ai/blog
- type: TermsOfService
  url: https://vast.ai/terms
- type: PrivacyPolicy
  url: https://vast.ai/privacy
- type: Support
  url: https://vast.ai/contact
- type: GitHubOrganization
  url: https://github.com/vast-ai
- type: LinkedIn
  url: https://www.linkedin.com/company/vast-ai
- name: vastai CLI
  url: https://github.com/vast-ai/vast-cli
  type: CLI
- name: Vast.ai Python SDK
  url: https://github.com/vast-ai/vast-sdk
  type: SDKs
- name: Vast.ai Base Image
  url: https://github.com/vast-ai/base-image
  type: Samples
- type: Features
  data:
  - name: GPU Marketplace
    description: Market-driven matching of buyers with 40+ datacenters and individual hosts across 68+ GPU types.
  - name: On-Demand Instances
    description: Per-second-billed GPU rentals with guaranteed uptime, with no minimum hours.
  - name: Interruptible Instances
    description: Fault-tolerant workloads priced 50%+ cheaper than on-demand with potential preemption.
  - name: Reserved Capacity
    description: Up to 50% discounts on 1, 3, or 6-month commitments for stable workloads.
  - name: Serverless Endpoints
    description: Serverless GPU workers built on the pyworker agent for autoscaling inference.
  - name: Templates
    description: Pre-configured instance templates for fast launching of common workloads.
- type: GPUs
  data:
  - name: NVIDIA B200
    description: Blackwell-generation flagship GPU available across hosts.
  - name: NVIDIA H200
    description: 141GB HBM3e GPU available across hosts.
  - name: NVIDIA H100
    description: 80GB Hopper GPU widely available across the marketplace.
  - name: NVIDIA A100
    description: 80GB Ampere GPU widely available across the marketplace.
  - name: NVIDIA RTX 4090
    description: 24GB consumer GPU, popular for inference and fine-tuning.
  - name: NVIDIA RTX 3060
    description: 12GB consumer GPU at entry-level price points.
- type: Integrations
  data:
  - name: Docker
    description: Bring-your-own Docker image support across all rental instances.
  - name: SSH
    description: Direct SSH access to rented instances for development and orchestration.
- type: LlmsText
  url: https://docs.vast.ai/llms.txt
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com
  url: https://apievangelist.com