specification: API Commons Rate Limits
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/RateLimits
provider: Toloka
providerId: toloka
created: '2026-06-21'
modified: '2026-06-21'
reconciled: false
tags:
- Data Labeling
- Crowdsourcing
- Human-in-the-Loop
- Training Data
- AI
- Rate Limiting
- Quotas
- Throttling
description: >-
  The Toloka API applies throughput limits primarily on bulk write operations.
  Task and task-suite uploads are capped at up to 100,000 items per minute and
  2,000,000 items per day, and large batches must be created with async_mode so
  the work is processed as an Operation that the client polls. Per-account
  request-rate limits also apply to read and write endpoints; specific
  per-endpoint values are not fully published and are not reconciled here.
notes: >-
  Use async_mode=true for large task/task-suite/bonus batches and poll the
  returned Operation rather than issuing many synchronous requests. Verify exact
  per-endpoint request rates and burst limits with Toloka support on
  reconciliation.
sources:
- https://toloka.ai/docs/api/
- https://toloka.ai/docs/api/concepts/upload-tasks
- https://github.com/Toloka/toloka-kit
responseCodes:
  throttled: 429
limits:
- name: Task Upload (per minute)
  scope: account
  metric: tasks
  limit: 100000
  notes: Maximum tasks created per minute via POST /tasks.
- name: Task Upload (per day)
  scope: account
  metric: tasks
  limit: 2000000
  notes: Maximum tasks created per day via POST /tasks.
- name: Task Suite Upload
  scope: account
  metric: task_suites
  limit: see provider documentation
  notes: Bulk task-suite creation; use async_mode for large batches.
- name: Request Rate
  scope: account
  metric: requests
  limit: see provider documentation
  notes: Per-account API request-rate limit; specific value not published.
policies:
- name: Asynchronous Batching
  description: Large task, task-suite, and bonus batches must use async_mode=true and be tracked via the Operations API.
- name: Backoff Strategy
  description: Clients should implement exponential backoff with jitter and honor Retry-After on 429 responses.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com