specification: API Commons Plans
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/Plans
provider: Anyscale
providerId: anyscale
created: '2026-05-08'
# Provenance stamped 2026-08-11: this artifact was written by the API Evangelist
# bulk sweep dated 2026-05-08, not harvested from the provider. See roadmap#35.
method: generated
modified: '2026-05-08'
reconciled: true
tags:
- AI
- Distributed Computing
- Ray
- ML Platform
- Inference
- GPU
- Plans
description: >-
  Anyscale uses a usage-based pay-as-you-go model billed in compute instance
  hours, with rates that vary by GPU type. Two delivery modes are offered -
  Hosted (monthly credit-card invoicing, business-hours support) and Bring
  Your Own Cloud (BYOC) (invoiced via cloud marketplaces, 24/7 support, can
  consume customer GPU reservations). New users receive $100 in free credits;
  committed contracts unlock volume discounts.
notes: >-
  Hosted hourly rates apply to Anyscale-managed compute. In BYOC, Anyscale
  charges a platform fee per instance hour on top of the customer's
  cloud-provider compute spend - exact BYOC rate is contract-dependent.
sources:
- https://www.anyscale.com/pricing
- https://docs.anyscale.com/
plans:
- id: anyscale-hosted
  name: Hosted (Pay-as-you-go)
  type: usage
  description: >-
    Anyscale-managed compute on AWS / GCP, billed per instance-hour by node
    type with monthly credit-card invoicing and business-hours support.
  entries:
  - label: CPU Only (per hour)
    name: cpu_only
    type: usage
    metric: hours
    limit: -1
    timeFrame: month
    geo: global
    unit: 1
    price: $0.0135 per hour
    userMultiplied: false
  - label: NVIDIA T4 (per hour)
    name: gpu_t4
    type: usage
    metric: hours
    limit: -1
    timeFrame: month
    geo: global
    unit: 1
    price: $0.5682 per hour
    userMultiplied: false
  - label: NVIDIA L4 (per hour)
    name: gpu_l4
    type: usage
    metric: hours
    limit: -1
    timeFrame: month
    geo: global
    unit: 1
    price: $0.9542 per hour
    userMultiplied: false
  - label: NVIDIA A10G (per hour)
    name: gpu_a10g
    type: usage
    metric: hours
    limit: -1
    timeFrame: month
    geo: global
    unit: 1
    price: $1.3635 per hour
    userMultiplied: false
  - label: NVIDIA A100 (per hour)
    name: gpu_a100
    type: usage
    metric: hours
    limit: -1
    timeFrame: month
    geo: global
    unit: 1
    price: $4.9591 per hour
    userMultiplied: false
  - label: NVIDIA H100 (per hour)
    name: gpu_h100
    type: usage
    metric: hours
    limit: -1
    timeFrame: month
    geo: global
    unit: 1
    price: $9.2880 per hour
    userMultiplied: false
  - label: NVIDIA H200 (per hour)
    name: gpu_h200
    type: usage
    metric: hours
    limit: -1
    timeFrame: month
    geo: global
    unit: 1
    price: $10.6812 per hour
    userMultiplied: false
  elements:
  - name: Workspaces
  - name: Jobs
  - name: Services
  - name: Clusters
- id: anyscale-byoc
  name: Bring Your Own Cloud (BYOC)
  type: enterprise
  description: >-
    Anyscale runs the control plane while compute lives in the customer's
    AWS / GCP account, leveraging existing GPU reservations. Invoiced via
    cloud marketplaces with 24/7 support.
  entries:
  - label: Anyscale Platform Fee
    name: byoc_platform_fee
    type: usage
    metric: hours
    limit: -1
    timeFrame: month
    geo: global
    unit: 1
    price: per instance-hour, contract-dependent (contact sales)
    userMultiplied: false
  - label: Underlying Cloud Compute
    name: byoc_cloud_compute
    type: usage
    metric: hours
    limit: -1
    timeFrame: month
    geo: global
    unit: 1
    price: billed by AWS / GCP at customer's rates
    userMultiplied: false
  elements:
  - name: 24/7 Support
  - name: GPU Reservation Reuse
  - name: Marketplace Billing
- id: anyscale-enterprise
  name: Enterprise / Committed
  type: enterprise
  description: >-
    Volume commitments unlock discounts on hosted hourly rates and BYOC
    platform fees. Contact Anyscale sales.
  entries:
  - label: Committed Spend
    name: enterprise
    type: flat
    metric: contract
    limit: -1
    timeFrame: year
    geo: global
    unit: 1
    price: contact sales (volume discount)
    userMultiplied: false
  elements:
  - name: Volume Discount
  - name: SLAs
  - name: Custom Onboarding
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com