apiCommonsRateLimits: 0.1
generated: '2026-08-13'
method: searched
source: https://www.lalal.ai/api/v1/docs/
limit_count: 0
provider: LALAL.AI
provider_id: lalal-ai
updated: '2026-05-25'
authority: https://www.lalal.ai/api/v1/docs/
notes: |
  LALAL.AI does not publish explicit per-key request-rate limits in its public API documentation. The enforced
  ceiling is processing minutes (consumed minutes = file_length * number_of_stems) and per-file upload size. Use
  POST /api/v1/limits/minutes_left/ to inspect remaining minutes for a license. Contact support@lalal.ai for
  dedicated quotas or higher concurrency on enterprise contracts.

quotas:
  - id: processing-minutes-fast-queue-lite
    plan: lite
    type: minutes_per_month
    amount: 90
    scope: Fast Queue
  - id: processing-minutes-fast-queue-pro
    plan: pro
    type: minutes_per_month
    amount: 250
    scope: Fast Queue
  - id: processing-minutes-relaxed-queue-paid
    plan: lite-or-pro
    type: minutes_per_month
    amount: unlimited
    scope: Relaxed Queue
  - id: processing-minutes-starter
    plan: starter
    type: minutes_lifetime
    amount: 10
    scope: Relaxed Queue (preview only)

upload_limits:
  - id: upload-size-starter
    plan: starter
    max_file_size_bytes: 209715200
    max_file_size_human: 200 MB
  - id: upload-size-paid
    plan: lite-or-pro
    max_file_size_bytes: 2147483648
    max_file_size_human: 2 GB
  - id: upload-size-api-license
    plan: licensed-api
    max_file_size_bytes: 10737418240
    max_file_size_human: 10 GB
    notes: Multi-part upload required for files above the single-PUT threshold.

retention:
  source_files_hours: 24
  description: Uploaded source files are deleted 24 hours after upload.

introspection:
  endpoint: POST /api/v1/limits/minutes_left/
  description: Returns remaining processing minutes for the calling license key.

runtime_signal:
  generated: '2026-08-13'
  method: probed
  source: >-
    https://www.lalal.ai/api/v1/openapi.json (declared responses) + a live unauthenticated POST to
    https://www.lalal.ai/api/v1/check/ on 2026-08-13
  request_rate_limits_published: 0
  exhaustion_status: 429
  exhaustion_declared_on:
    - POST /api/v1/check/
  exhaustion_schema: CommonErrorResponse
  response_headers:
    ratelimit_standard: absent
    x_ratelimit: absent
    retry_after: absent
  observed_response_headers:
    - x-request-id
    - vary
    - cache-control
    - access-control-allow-origin
    - access-control-allow-methods
    - access-control-allow-headers
  note: >-
    429 Too Many Requests is declared in the contract on the polling endpoint only, and no RateLimit-*,
    X-RateLimit-* or Retry-After header is documented or was observed on a live response. There is no
    runtime budget signal an agent can read — a client must back off exponentially with no server hint.
    The only introspectable ceiling is the minute balance from POST /api/v1/limits/minutes_left/, which
    reports remaining quota but not remaining request rate.
  quota_introspection:
    endpoint: POST /api/v1/limits/minutes_left/
    operation_id: limits_get_minutes_left
    field: minutes_left
    type: number