name: Smithsonian Open Access API FinOps
description: Financial operations profile for the Smithsonian Open Access API. The
  API itself is free with no direct monetary costs. Cost considerations are limited
  to downstream infrastructure when consuming bulk data via AWS S3 or building applications
  on top of the API.
url: https://www.si.edu/openaccess
created: '2026-06-13'
modified: '2026-06-13'
costModel:
  type: Free / Public Good
  description: The Smithsonian Open Access API carries no direct cost to consumers.
    It is funded by the Smithsonian Institution (a US federal trust instrumentality)
    and operated as a public service. All content is released under CC0 (public domain).
  apiCost: $0.00
  registrationCost: $0.00
  dataLicensingCost: $0.00
directCosts:
  - category: API Access
    cost: Free
    notes: No charges for API calls regardless of volume (within rate limits)
  - category: API Key Registration
    cost: Free
    notes: Registration via api.data.gov requires no payment information
  - category: Content Licensing
    cost: Free
    notes: All Open Access content is CC0 (public domain) with no licensing fees
indirectCosts:
  - category: AWS S3 Bulk Data Egress
    description: When accessing bulk data via the public S3 bucket (s3://smithsonian-open-access),
      AWS may charge standard data transfer fees on the consumer's AWS account for
      data downloaded out of us-west-2.
    estimatedCost: AWS standard egress rates (varies by volume and destination)
    mitigation: Use AWS services in us-west-2 to minimize inter-region transfer costs
  - category: Developer Time (Integration)
    description: Engineering effort to integrate with the API and process collection
      metadata in various formats (JSON, XML, CSV, RDF).
    estimatedCost: Variable — API is well-documented with Python SDK available (si_openaccess)
  - category: Storage (Bulk Data)
    description: If mirroring or caching the S3 bulk dataset (~2.8M+ CC0 images),
      storage costs apply on the consumer's infrastructure.
    estimatedCost: Variable — dataset is multi-terabyte and grows weekly
  - category: Compute
    description: Processing and indexing 4.7M+ collection records requires compute
      resources on the consumer side.
    estimatedCost: Variable
budgetGuidance:
  smallTeams: No budget required for standard API access. Register at api.data.gov
    for a free key and start building immediately. Budget only for engineering time.
  largeOrganizations: For bulk data ingestion via S3, plan for AWS egress costs and
    storage. Consider running workloads in us-west-2 (same region as the S3 bucket)
    to reduce transfer costs. Contact openaccess@si.edu for high-volume API needs.
  costOptimization:
    - Use the S3 bulk dataset for initial data loads; use the API for incremental updates
    - The S3 dataset updates weekly — sync incrementally rather than full re-download
    - Cache API responses locally to stay within the 1,000 requests/hour limit
    - Use the DEMO_KEY only for prototyping; register a full key before production use
    - Use search_terms endpoint to pre-filter queries and reduce unnecessary API calls
openSourceTools:
  - name: si_openaccess Python SDK
    url: https://pypi.org/project/si-openaccess/
    repo: https://github.com/Smithsonian/smithsonian-openaccess
    cost: Free (Apache 2.0)
    description: Official Python module for querying the Smithsonian Open Access API
  - name: Smithsonian Open Access MCP Server
    url: https://github.com/molanojustin/smithsonian-mcp
    cost: Free
    description: MCP server enabling AI assistant access to the Smithsonian collections
fundingModel:
  provider: Smithsonian Institution (US federal trust instrumentality)
  description: The Smithsonian Institution is funded through a combination of federal
    appropriations and private donations. Open Access digitization and API infrastructure
    is part of its public mission.
  contact: openaccess@si.edu