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: Smallest AI
providerId: smallest-ai
created: '2026-06-21'
modified: '2026-06-21'
reconciled: false
tags:
- AI
- Text to Speech
- Voice
- Realtime
- Voice Agents
- Rate Limiting
- Quotas
- Throttling
description: >-
  The Smallest AI Waves API enforces per-account limits on requests per minute
  (RPM) and on the number of concurrent realtime/streaming connections. On the
  pay-as-you-go tier these are documented around 100 RPM and 15 concurrent
  streams, with higher limits available on Enterprise. Specific per-model and
  per-tier values are not reconciled in this artifact.
notes: >-
  Verify per-model RPM and concurrent-stream limits on the Smallest AI pricing
  and documentation pages during reconciliation; limits change between pay-as-
  you-go, subscription, and Enterprise tiers.
sources:
- https://smallest.ai/pricing
- https://docs.smallest.ai/
responseCodes:
  throttled: 429
limits:
- name: Requests Per Minute (RPM)
  scope: account
  metric: requests
  limit: ~100 (pay-as-you-go)
  notes: Per-account RPM on the TTS API; varies by tier.
- name: Concurrent Streams
  scope: account
  metric: connections
  limit: ~15 (pay-as-you-go)
  notes: Maximum simultaneous SSE/WebSocket realtime TTS connections; varies by tier.
- name: Concurrent Requests
  scope: account
  metric: connections
  limit: see provider documentation
  notes: Applies to synchronous get_speech requests; varies by tier.
policies:
- name: Tiered Limits
  description: Limits raise from pay-as-you-go to subscription and Enterprise agreements.
- name: Backoff Strategy
  description: Clients should implement exponential backoff with jitter and honor Retry-After on 429.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com