name: APIMatic API Rate Limits 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: APIMatic providerId: apimatic created: '2026-05-04' modified: '2026-09-15' generated: '2026-09-15' method: searched source: |- https://www.apimatic.io/pricing ; https://docs.apimatic.io/ (full text searched via https://docs.apimatic.io/llms-full.txt) ; https://github.com/apimatic/apimatic-sdk-for-js/tree/main/doc/controllers note: >- Replaces a 2026-05-04 scaffold that asserted per-minute and per-month request limits, burst ceilings and X-RateLimit-* response headers APIMatic does not publish. THE HONEST FINDING: APIMatic documents NO per-request rate limit, NO throttling window, NO 429 contract and NO rate-limit response headers for its own Platform API. A full-text search of the 954 KB llms-full.txt returns zero occurrences of "X-RateLimit"; every hit on 429 and Retry-After is about the SDKs APIMatic GENERATES FOR CUSTOMERS, not about its own endpoints. What APIMatic does publish are plan entitlement ceilings, and those are recorded below. tags: - API Transformation - Code Generation - Developer Experience - Quotas description: >- Published limits for the APIMatic surface. All of them are plan entitlement ceilings taken from the pricing page; none is a request-rate limit. Runtime rate-limit signalling is absent. headers: limit: null remaining: null reset: null retryAfter: null policy: null documented: false responseCodes: throttled: null quotaExceeded: 403 quotaExceededNote: >- Subscription and entitlement failures surface as HTTP 403 "Subscription Issue" on the validation, code-generation and portal-management controllers — the only documented response to exceeding what a plan includes. documented: partial limits: - tier: starter name: Endpoints per API (Lite/Starter) scope: account-plan metric: endpoints_per_api limit: 20 timeFrame: perpetual source: https://www.apimatic.io/pricing - tier: starter name: API portals (Lite/Starter) scope: account-plan metric: portals limit: 1 timeFrame: perpetual source: https://www.apimatic.io/pricing - tier: starter name: SDK languages (Lite/Starter) scope: account-plan metric: sdk_languages limit: 1 timeFrame: perpetual source: https://www.apimatic.io/pricing - tier: basic name: SDK languages (Basic) scope: account-plan metric: sdk_languages limit: 5 timeFrame: perpetual source: https://www.apimatic.io/pricing - tier: business name: Endpoints per API (Business) scope: account-plan metric: endpoints_per_api limit: 100 timeFrame: perpetual source: https://www.apimatic.io/pricing - tier: business name: SDK languages (Business) scope: account-plan metric: sdk_languages limit: 7 timeFrame: perpetual source: https://www.apimatic.io/pricing - tier: platform name: API entity name length scope: per-request metric: characters limit: 100 timeFrame: perpetual source: https://docs.apimatic.io/changelog/added-max-name-length-validation-rules note: >- Validation and linting raise an Error-severity violation when an API entity name exceeds 100 characters. A real published hard limit, enforced at import/validation time. limit_count: 7 request_rate_limits_documented: false policies: - name: SDK-side retry guidance (customer APIs, not APIMatic's own) description: >- APIMatic's generated SDKs retry on 408, 413, 429, 500, 502, 503, 504, 521, 522 and 524 with exponential backoff and honour Retry-After. This is the retry contract APIMatic ships to its customers' consumers; it is not a statement about APIMatic's own endpoints. source: https://docs.apimatic.io/generate-sdks/sdk-features/retries-with-exponential-backoff/ gaps: - >- No per-key or per-account request rate limit is published for https://api.apimatic.io. - >- No X-RateLimit-* / RateLimit-* response headers and no documented 429 behaviour, so an agent has no runtime signal to back off on. maintainers: - FN: Kin Lane email: kin@apievangelist.com