specification: API Commons Rate Limits
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/RateLimits
provider: First Street
providerId: first-street
generated: '2026-09-10'
method: searched
source: https://docs.firststreet.org/api/climate-risk-api/getting-started/rate-limits
created: '2026-05-04'
modified: '2026-09-10'
tags:
  - Environment
  - Modeling
  - Risk
  - Climate
  - Rate Limiting
  - Quotas
  - Throttling
description: >-
  First Street's published rate limits, replacing the 2026-05-04 scaffold that carried
  invented free/pro/enterprise tiers. The real contract has ONE published default —
  150 requests per minute — and everything else is negotiated. It also carries a second,
  independent ceiling most catalogs miss: a query-complexity budget of 700 that can reject
  a single request no matter how slowly you send them.
limit_count: 2

headers:
  limit: x-ratelimit-limit
  remaining: x-ratelimit-remaining
  reset: x-ratelimit-reset
  retryAfter: null
header_semantics:
  x-ratelimit-limit: The max number of requests available during the cycle (the rate limit).
  x-ratelimit-remaining: The amount of requests remaining in the current 1 minute cycle.
  x-ratelimit-reset: The number of seconds until remaining limits reset.
  note: >-
    No Retry-After and no RateLimit-Policy is documented; x-ratelimit-reset is the only
    backoff signal an agent gets.

responseCodes:
  throttled: 429

limits:
  - name: Default request limit
    scope: api-key
    metric: requests_per_minute
    limit: 150
    window: minute
    burst: null
    tier: default
    negotiable: true
    note: >-
      "Default request limiting starts at 150 requests per minute but is adjusted based on
      individual contract agreements."
    applies:
      - First Street Climate Risk API
      - First Street Enterprise API
      - First Street Raster Map API

  - name: Query complexity budget
    scope: request
    metric: complexity_score
    limit: 700
    window: per-request
    tier: default
    negotiable: false
    note: >-
      "Query complexity is based off of the data fields requested in a query ... The
      current query complexity limit is set to 700 which should satisfy 99% of all user
      requests." A second, orthogonal ceiling: a wide GraphQL selection can be rejected
      even when the caller is well inside the per-minute limit. The scoring algorithm is
      internal and not published, so the cost of a given query cannot be computed client
      side.
    applies:
      - First Street Climate Risk API
      - First Street Enterprise API

service_specific:
  supported: true
  note: >-
    "We may apply isolated rate limits to individual services when requested. For example,
    a limit of 150 requests per limit may be applied to our GraphQL API while a limit of
    1000/requests per limit may be applied to our Tiles API. These limits will be based on
    contractual agreements and applied during key retrevial."
  example_tiles_limit: 1000 requests per minute (illustrative, contractual)

policies:
  - name: Contractual adjustment
    description: >-
      Every published number is a starting point. Actual limits are set per contract and
      bound to the key at issuance, so two keys against the same endpoint can have
      different ceilings.
  - name: Playground counts against usage
    description: >-
      Requests made through the GraphiQL Playground are ordinary API requests and count
      toward account usage; there is no free evaluation lane.
  - name: Tile usage counts
    description: >-
      Map tile requests, including those made from the published Mapbox/Google Maps
      examples, count toward overall API usage.

not_published:
  - A monthly or annual request quota
  - A burst allowance
  - Per-endpoint limits (offered on request, never published as numbers)
  - Retry-After on 429

maintainers:
  - FN: Kin Lane
    email: kin@apievangelist.com