specification: API Commons FinOps 0.1
name: Daytona FinOps Profile
description: >-
  Aligns the Daytona Cloud billing surface to the FinOps Framework / FOCUS data spec. Captures the resource
  dimensions billed (vCPU, memory, storage, GPU), pricing model (per-second, pay-as-you-go), free allowances,
  cost attribution tags, and the APIs and dashboards available for cost observability.
url: https://raw.githubusercontent.com/api-evangelist/daytona-io/refs/heads/main/finops/daytona-io-finops.yml
provider: daytona-io
humanURL: https://www.daytona.io/pricing
created: '2026-05-25'
modified: '2026-05-25'

billing:
  model: pay-as-you-go
  granularity: per-second
  currency: USD
  invoicing: monthly
  minimumCommit: 0
  signupCredit:
    amount: 200
    currency: USD
    description: Automatic $200 credit applied on signup.
  freeAllowance:
  - resource: storage
    amount: 5
    unit: GiB
    description: First 5 GiB of persistent storage free per organization.

resources:
- name: vCPU
  unit: vCPU-hour
  ratePerHour: 0.0504
  ratePerSecond: 0.000014
  focusServiceCategory: Compute
  focusChargeCategory: Usage
  description: Per-vCPU billing for running sandboxes.
- name: Memory
  unit: GiB-hour
  ratePerHour: 0.0162
  ratePerSecond: 0.0000045
  focusServiceCategory: Compute
  focusChargeCategory: Usage
  description: Per-GiB memory billing for running sandboxes.
- name: Storage
  unit: GiB-hour
  ratePerHour: 0.000108
  ratePerSecond: 0.00000003
  focusServiceCategory: Storage
  focusChargeCategory: Usage
  description: Persistent storage billed after the first 5 GiB free.
- name: GPU H100
  unit: GPU-hour
  ratePerHour: 3.95
  ratePerSecond: 0.001097
  focusServiceCategory: AI and Machine Learning
  focusChargeCategory: Usage
  description: Nvidia H100 GPU attached to GPU-class sandboxes.
- name: GPU RTX PRO 6000
  unit: GPU-hour
  ratePerHour: 3.03
  ratePerSecond: 0.000842
  focusServiceCategory: AI and Machine Learning
  focusChargeCategory: Usage
  description: Nvidia RTX PRO 6000 GPU attached to GPU-class sandboxes.

costAttribution:
  dimensions:
  - name: Organization
    description: Top-level billing boundary; aligns to FOCUS BillingAccountId.
  - name: Sandbox Labels
    description: Free-form key/value labels on sandboxes; map to FOCUS Tags for chargeback and showback.
  - name: Region
    description: Sandbox region; aligns to FOCUS Region.
  - name: Sandbox ID
    description: Per-sandbox cost attribution; aligns to FOCUS ResourceId.
  - name: Snapshot ID
    description: Per-snapshot storage cost attribution.
  - name: Volume ID
    description: Per-volume storage cost attribution.

observability:
  dashboard:
    url: https://app.daytona.io/
    description: Daytona Dashboard surfaces current usage, spend, and per-resource breakdowns.
  apis:
  - url: https://www.daytona.io/docs/en/account-management/billing
    description: Billing documentation including invoice access and payment management.
  - url: https://www.daytona.io/docs/en/observability/opentelemetry-collection
    description: OpenTelemetry collection for emitting sandbox traces and metrics to customer-owned backends.
  exports:
  - format: CSV
    description: Invoice export available from the billing section of the dashboard.

programs:
- name: Startups Program
  description: Up to $50k in free Daytona compute credits for qualifying startups.
  url: https://www.daytona.io/pricing
- name: Enterprise
  description: Custom-priced contracts with optional Customer-Managed Compute or self-hosted open-source.
  url: https://www.daytona.io/contact

recommendations:
- audience: ai-agent-platforms
  note: >-
    Use organization-level API keys and sandbox labels to attribute cost to individual tenants in
    multi-tenant agent platforms.
- audience: enterprises
  note: >-
    Evaluate Customer-Managed Compute for predictable cost under heavy use; pair with reserved GPU pools to
    reduce per-second GPU spend.
- audience: researchers
  note: >-
    Apply for the Startups Program early; $50k of credits can fund a large SWE-bench-style evaluation run.