generated: '2026-08-13'
method: derived
source: openapi/blng-journey-api-openapi.yml, openapi/blng-user-api-openapi.yml
docs: https://docs.blng.ai/account-management/usage-limits-for-free-plan/
note: >-
  BLNG publishes no rate-limit documentation and no developer program, so there is no numeric
  requests-per-window limit to record. What IS specified, in the specs themselves, is a per-user
  CONCURRENCY limit on generation and a per-invitation resend cooldown, both surfaced as HTTP 429 with
  a JSON body. No RateLimit-*, X-RateLimit-* or Retry-After response header is declared anywhere in the
  three specs, so a client has no runtime signal to back off on beyond the status code and the code
  string in the body. The only published consumption limits are plan quotas (files, history retention),
  which are recorded in plans/blng-plans-pricing.yml, not here.
limit_count: 2
headers_published: false
response_headers: []
exhaustion_status: 429
error_envelope:
  fields:
  - code
  - message
  example_code: CONCURRENCY_LIMIT_EXCEEDED
limits:
- name: Chat prompt concurrency
  scope: per-user
  window: concurrent
  limit: null
  limit_note: >-
    Value is not published. The spec states only that a user has "a concurrency limit for chat prompts"
    and returns 429 with code CONCURRENCY_LIMIT_EXCEEDED when it is exceeded. The pricing page markets
    "Prompt Concurrency" and "Priority queue" as tier features, implying the ceiling varies by plan.
  burst: null
  status: 429
  code: CONCURRENCY_LIMIT_EXCEEDED
  retry_after_header: false
  operations:
  - submitChatPrompt
  - submitChatPromptV3
  paths:
  - POST /design/journeys/{journeyId}/chat-prompt
  - POST /design/journeys/{journeyId}/chat-prompt-v3
  source: openapi/blng-journey-api-openapi.yml
- name: Workspace invitation resend cooldown
  scope: per-invitation
  window: cooldown
  limit: null
  limit_note: >-
    Cooldown duration is not published. The spec describes 429 as "Too many resend attempts for this
    invitation (cooldown between successful invitation emails)".
  burst: null
  status: 429
  retry_after_header: false
  paths:
  - POST /users/{userId}/workspaces/{workspaceId}/invitations/{invitationId}/resend
  source: openapi/blng-user-api-openapi.yml
related_backpressure:
- name: Journey operation lock
  status: 423
  description: >-
    Not a rate limit but the adjacent backpressure signal an agent must handle: starting a 3D model
    generation while a chat prompt (or another generation) is in flight returns 423 Locked with
    operationInFlight, operationOwnerId and operationExpiresAt, so the client knows when the lock frees.
  paths:
  - POST /design/journeys/{journeyId}/generations
  source: openapi/blng-journey-api-openapi.yml
- name: DynamoDB throughput
  status: 503
  description: Accepting a workspace invitation may return 503 on DynamoDB throughput limit; the spec instructs the client to retry.
  paths:
  - POST /users/{userId}/workspace-invitations/{invitationId}/accept
  source: openapi/blng-user-api-openapi.yml
x-evidence:
- url: https://docs.blng.ai/account-management/usage-limits-for-free-plan/
  http_status: 200
  fetched: '2026-08-13'
  finding: plan-level usage limits (3 active files, 24 hour history) — no API rate limits documented