generated: '2026-09-09'
method: searched
source: https://transcriptfetch.com/docs/rate-limits
docs: https://transcriptfetch.com/docs/rate-limits
limit_count: 5
scope_note: >-
  Limits are applied per API key (dashboard/session calls per user) over a sliding
  60-second window; each key can carry its own custom limit, and higher limits are
  available via support.
limits:
- scope: per-key
  plan: Free
  window: 60s sliding
  limit: 10 requests/minute
  source: https://transcriptfetch.com/pricing (rate-limit FAQ) and https://transcriptfetch.com/llms.txt
- scope: per-key
  plan: Basic
  window: 60s sliding
  limit: 60 requests/minute
- scope: per-key
  plan: Pro
  window: 60s sliding
  limit: 300 requests/minute
- scope: per-key
  plan: Mega
  window: 60s sliding
  limit: 900 requests/minute
- scope: per-key
  plan: Scale
  window: 60s sliding
  limit: 900 requests/minute (custom limits available; Enterprise offers custom rate limits)
headers:
- X-RateLimit-Limit
- X-RateLimit-Remaining
- X-RateLimit-Reset
- Retry-After
exhaustion:
  status: 429
  error_code: rate_limited
  error_number: 2101
  behavior: >-
    Body is the standard error envelope; Retry-After carries seconds to wait. A 429
    never consumes credits and always means the caller's own key limit — upstream
    platform blocks answer 503 upstream_error (5001) instead.