generated: '2026-08-25'
method: searched
source: >-
  Full-text search of the ModelOp Documentation Hub V3.4 (https://modelopdocs.atlassian.net/wiki/spaces/MDHV)
  for "rate limit", "throttle", "quota", "429" and "Retry-After"; plus https://www.modelop.com/llms.txt
checked: '2026-08-25'
limit_count: 0
summary: >-
  ModelOp publishes NO API rate limits. A full-text search of the current documentation space
  returns three hits for "rate limit" and all three are about the rate limits of THIRD-PARTY
  systems ModelOp integrates with (ServiceNow, Jira) or an unrelated release note — none describes
  a limit ModelOp Center itself enforces. No 429 status, no Retry-After, and no X-RateLimit-* or
  RateLimit-* response header is documented anywhere.
rate_limits: []
response_headers:
  documented: []
  note: >-
    No rate-limit response headers are documented. An agent has no runtime signal to back off on.
exhaustion_status_code: null
structural_note: >-
  ModelOp Center runs inside the customer's own Kubernetes cluster, so the practical limits are
  whatever the customer's own sizing allows rather than a vendor-enforced quota. The docs do carry
  a "ModelOp Center Sizing Considerations" page
  (https://modelopdocs.atlassian.net/wiki/spaces/MDHV/pages/3159993300/ModelOp+Center+Sizing+Considerations)
  which is a capacity-planning guide, NOT a rate-limit contract, and it is not counted here.
adjacent_documented_limits:
  - limit: 10 MB
    scope: model asset upload
    behaviour: >-
      Assets over 10 MB must be stored in configured external storage rather than inside
      model-manage (`moc asset add --external`, or `--force` to choose automatically).
    source: https://modelopdocs.atlassian.net/wiki/spaces/MDHV/pages/3159993475/ModelOp+CLI+Reference
    note: A payload-size threshold, not a request-rate limit. Recorded so it is not mistaken for one.
  - limit: 5000 ms
    scope: ModelOp Runtime round-trip REST scoring call
    behaviour: Default timeout for a synchronous round-trip scoring request; configurable via modelop.runtime.roundtrip-default-timeout.
    source: https://modelopdocs.atlassian.net/wiki/spaces/MDHV/pages/3159997056/ModelOp+Runtime+Details
    note: A timeout, not a rate limit.
third_party_rate_limits_referenced:
  - system: ServiceNow
    source: https://modelopdocs.atlassian.net/wiki/spaces/MDHV/pages/3159993730/Integrate+with+ServiceNow
  - system: Jira
    source: https://modelopdocs.atlassian.net/wiki/spaces/MDHV/pages/3159993648/Integrate+with+JIRA
note: >-
  An honest zero — recorded because a measured absence is data. limit_count: 0 with this evidence
  is a different fact from "not checked".