generated: '2026-07-19'
method: searched
source: https://docs.landing.ai/dpt3/rate-limits.md, https://docs.landing.ai/ade/ade-rate-limits.md
description: >-
  ADE rate limiting is metered on pages processed against a plan-level hourly
  limit at the organization level, not on request counts. Hard size and page
  ceilings differ sharply between the synchronous APIs and the asynchronous Jobs
  APIs.
basis: pages processed per hour, per organization, by plan
signaling:
  status_code: 429
  headers: null
  guidance: Wait before retrying and use exponential backoff.
  async_behavior: >-
    Job submissions are accepted immediately (HTTP 202) and queued rather than
    rejected, which is the documented way to avoid rate-limit errors on bulk work.
rate_limits:
- api: Parse v2 (synchronous)
  limit_type: document_size
  limit_count: 50
  limit_unit: MiB
- api: Parse v2 (synchronous)
  limit_type: pages_per_request
  limit_count: 100
  limit_unit: pages per PDF
  note: An image counts as one page.
- api: Parse Jobs v2 (asynchronous)
  limit_type: document_size
  limit_count: 1
  limit_unit: GiB
  note: 1 GiB for PDFs, 50 MiB for images.
- api: Parse Jobs v2 (asynchronous)
  limit_type: pages_per_request
  limit_count: 6000
  limit_unit: pages per PDF
- api: Extract v2 (synchronous)
  limit_type: input_size
  limit_count: null
  limit_unit: unlimited
  note: Rate limit is based on input Markdown size; a markdown_url counts as a single unit.
- api: Extract Jobs v2 (asynchronous)
  limit_type: input_size
  limit_count: null
  limit_unit: unlimited
plan_tiers:
- plan: Explore
  rate_limits: standard
- plan: Team
  rate_limits: higher
- plan: Enterprise
  rate_limits: priority
credits:
  docs: https://docs.landing.ai/dpt3/credit-consumption
  consumed_on: [200, 206]
  free_on_error: true
  note: Errors consume no credits.