generated: '2026-08-13'
method: searched
source: https://developers.line.biz/en/reference/messaging-api/#rate-limits
docs: https://developers.line.biz/en/reference/messaging-api/
summary:
  limit_count: 9
  scope: per-endpoint, per-channel
  algorithm: token bucket
  exhaustion_status: 429
  response_headers_published: false
  retry_after: false
  note: >-
    LINE documents a per-endpoint rate limit table but returns no rate-limit
    response headers of any kind — no RateLimit-*, no X-RateLimit-*, no
    Retry-After. An agent gets no runtime budget signal; it learns it is over
    the line only by receiving 429 Too Many Requests. This is the single
    largest runtime-semantics gap on the platform.
scope_rules:
  - Limits apply per API function (endpoint) on a per-channel basis.
  - Same URL with a different HTTP method counts as a different endpoint.
  - >-
    Limits ignore URL parameter values and request body contents — all calls to
    one endpoint share one bucket.
  - Limits ignore source IP address.
  - >-
    Two channels attached to the same LINE Official Account are limited
    independently.
limits:
  - scope: per-channel-per-endpoint
    window: 1 hour
    limit: 60
    endpoints:
      - narrowcast
      - broadcast
      - getNumberOfMessageDeliveries
      - getNumberOfFollowers
      - getFriendsDemographics
      - getMessageEvent
      - getStatisticsPerUnit
      - getRichMenuInsightSummary
      - getRichMenuInsightDaily
      - testWebhookEndpoint
  - scope: per-channel-per-endpoint
    window: 1 minute
    limit: 60
    endpoints:
      - createAudienceGroup
      - createAudienceForUploadingUserIds
      - addAudienceToAudienceGroup
      - addUserIdsToAudience
      - createClickBasedAudienceGroup
      - createImpBasedAudienceGroup
      - updateAudienceGroupDescription
      - deleteAudienceGroup
      - getAudienceData
      - getAudienceGroups
      - getSharedAudienceData
      - getSharedAudienceGroups
  - scope: per-channel-per-endpoint
    window: 1 minute
    limit: 1000
    endpoints:
      - setWebhookEndpoint
      - getWebhookEndpoint
  - scope: per-channel-per-endpoint
    window: 1 hour
    limit: 100
    endpoints:
      - createRichMenu
      - deleteRichMenu
      - deleteRichMenuAlias
      - getRichMenuBatchProgress
    note: >-
      Creating and deleting rich menus through the LINE Official Account
      Manager UI is not counted against this limit.
  - scope: per-channel-per-endpoint
    window: 1 hour
    limit: 3
    endpoints:
      - richMenuBatch
  - scope: per-channel-per-endpoint
    window: 1 second
    limit: 10
    endpoints:
      - getRichMenuList
  - scope: per-channel-per-endpoint
    window: 1 second
    limit: 200
    endpoints:
      - multicast
      - getMembershipSubscription
      - getMembershipList
      - createCoupon
      - closeCoupon
      - listCoupon
      - getCouponDetail
  - scope: per-channel-per-endpoint
    window: 1 second
    limit: 100
    endpoints:
      - showLoadingAnimation
  - scope: per-channel-per-endpoint
    window: 1 second
    limit: 370
    endpoints:
      - issueStatelessChannelToken
  - scope: per-channel-per-endpoint
    window: 1 second
    limit: 2000
    endpoints:
      - all other endpoints
    note: Default limit for every Messaging API endpoint not named above.
concurrency_limits:
  - scope: per audienceGroupId
    max_concurrent: 10
    operations:
      - createAudienceGroup
      - createAudienceForUploadingUserIds
      - addAudienceToAudienceGroup
      - addUserIdsToAudience
    observable_via: >-
      getAudienceData — jobs[].jobStatus of QUEUED or WORKING counts toward the
      concurrency total.
    on_exceed: 429
quota_limits:
  - name: monthly message quota
    scope: per LINE Official Account subscription plan
    observable_via: [getMessageQuota, getMessageQuotaConsumption]
    on_exceed: 429 with message "You have reached your monthly limit."
    note: >-
      A 429 on LINE can mean rate limit, concurrency limit, OR monthly message
      quota exhaustion — three different conditions on one status code,
      distinguishable only by the `message` string in the body.
    see: plans/line-plans-pricing.yml