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: Hypertune
providerId: hypertune
created: '2026-06-20'
modified: '2026-06-20'
reconciled: false
tags:
- Feature Flags
- Experimentation
- Analytics
- Edge
- Rate Limiting
- Quotas
- Throttling
description: >-
  Hypertune does not publish numeric per-request rate limits for the Edge
  GraphQL API or SDK initialization. Instead, plan capacity is governed by
  metered quotas: included CDN requests (SDK initialization / GraphQL Edge
  evaluation) and analytics events per month, with per-million overage billing
  on paid tiers. Because SDKs fetch flag logic once and then evaluate flags
  locally in memory, the dominant request volume is initialization and
  background log flushing rather than per-flag-evaluation calls.
notes: >-
  No explicit RPM/RPS throttling thresholds are documented; capacity is
  expressed as monthly CDN-request and analytics-event quotas on the pricing
  page. Verify current overage rates and any soft limits with Hypertune on
  reconciliation. NOTE: Hypertune is shutting down on August 10, 2026.
sources:
- https://www.hypertune.com/pricing
- https://docs.hypertune.com/pricing/overview
- https://docs.hypertune.com/concepts/architecture
responseCodes:
  throttled: 429
limits:
- name: CDN Requests (Free)
  scope: account
  metric: requests
  limit: 1000000
  notes: Included SDK initialization / GraphQL Edge requests per month on the Free plan.
- name: CDN Requests (Starter / Pro)
  scope: account
  metric: requests
  limit: 2000000
  notes: 2M/month included, then $10 per additional 1M.
- name: Analytics Events (Free)
  scope: account
  metric: events
  limit: 1000000
  notes: Included analytics events per month on the Free plan.
- name: Analytics Events (Starter / Pro)
  scope: account
  metric: events
  limit: 2000000
  notes: 2M/month included, then $30 per additional 1M.
- name: Enterprise Volume
  scope: account
  metric: requests
  limit: see provider documentation
  notes: Flexible volume-based and MTU pricing negotiated per contract.
policies:
- name: Metered Quotas
  description: Capacity is governed by monthly CDN-request and analytics-event quotas with per-million overage, not fixed RPM/RPS throttles.
- name: Local Evaluation
  description: SDKs fetch flag logic once and evaluate locally in memory, minimizing per-evaluation network requests.
- name: Background Flushing
  description: Analytics events and exposures are batched and flushed to Hypertune Edge in the background to reduce request volume.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com