generated: '2026-09-19'
method: searched
source: https://relmcrm.com/docs (Rate limits & quotas) + https://relmcrm.com/errors (rate_limited, quota_exceeded, spend_cap_reached)
checked: '2026-09-19'
corroboration: >-
  Live anonymous probes on 2026-09-19 (GET https://api.relmcrm.com/v1/schema -> 401 problem+json; POST
  https://api.relmcrm.com/mcp tools/call -> 401) returned no X-RateLimit-* or X-Quota-* headers, which is
  consistent with the docs: the headers ride on authenticated responses, and this pass holds no key. Header
  NAMES below come from the docs; no numeric value has been observed live.
confidence: medium
confidence_note: >-
  The per-minute rate limit is documented to exist but its NUMBER is not published anywhere (docs, llms.txt,
  pricing, error reference); Scale advertises "top-tier rate limits" without a figure. The monthly quotas are
  published exactly. Treat the header family as authoritative for client design and read the live limit from
  X-RateLimit-* on the first authenticated response.
limit_count: 4
scopes:
- scope: per-workspace
  surface: REST (api.relmcrm.com/v1) + MCP (api.relmcrm.com/mcp) + A2A — one workspace budget
  window: 1 minute
  limit: null
  limit_note: Documented as "rate-limited per workspace per minute"; the number is not published.
  unit: requests
  burst: null
  status_on_exhaustion: 429
  error_code: rate_limited
  headers:
  - name: X-RateLimit-*
    meaning: Per-minute rate-limit family named in the docs ("Responses carry X-RateLimit-* and X-Quota-* headers"); individual member names are not enumerated.
  - name: Retry-After
    meaning: Named on the rate_limited error ("Back off and retry (see Retry-After)").
  retry_after: documented
  source: https://relmcrm.com/docs
- scope: per-workspace
  surface: all surfaces — Free plan
  window: monthly (billing period)
  limit: 1000
  unit: requests
  status_on_exhaustion: 429
  error_code: quota_exceeded
  headers:
  - name: X-Quota-*
    meaning: Monthly-quota family named in the docs; member names not enumerated. GET /v1/usage returns used/limit/overage/plan.
  note: Hard stop at the cap; no overage on Free.
  source: https://relmcrm.com/docs
- scope: per-workspace
  surface: all surfaces — Pro plan
  window: monthly (billing period)
  limit: 100000
  unit: requests
  status_on_exhaustion: 429 only when the customer-set spend cap is hit (spend_cap_reached); otherwise metered overage at $0.0001/request
  error_code: spend_cap_reached
  source: https://relmcrm.com/docs
- scope: per-workspace
  surface: all surfaces — Scale plan
  window: monthly (billing period)
  limit: 2000000
  unit: requests
  status_on_exhaustion: 429 only when the customer-set spend cap is hit (spend_cap_reached); otherwise metered overage at $0.0001/request
  error_code: spend_cap_reached
  source: https://relmcrm.com/docs
plan_caps:
- scope: per-workspace — Free plan
  resource: automations
  limit: 2
  status_on_exhaustion: 403
  error_code: plan_limit
- scope: per-workspace — Free plan
  resource: sequences
  limit: 1
  status_on_exhaustion: 403
  error_code: plan_limit
metering:
  test_mode_counts: false
  batch: metered per operation (POST /v1/batch up to 100 operations)
  mcp_array_post: metered per tool call
  spend_cap: 'PATCH /v1/settings {"overage_cap_usd": <number|null>} (relm_set_spend_cap); $0 hard-stops like Free'
exhaustion_codes:
- status: 429
  code: rate_limited
  meaning: Too many requests this minute. Back off and retry (see Retry-After).
- status: 429
  code: quota_exceeded
  meaning: The monthly request quota was reached. Check GET /v1/usage; upgrade to continue.
- status: 429
  code: spend_cap_reached
  meaning: 'The monthly overage spend cap was reached. Raise or clear it: PATCH /v1/settings {"overage_cap_usd": null}.'
- status: 403
  code: plan_limit
  meaning: A plan cap was reached (e.g. Free allows 2 automations / 1 sequence).