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: Scrapybara
providerId: scrapybara
created: '2026-07-01'
modified: '2026-07-01'
reconciled: false
tags:
- AI Agents
- Virtual Desktops
- Computer Use
- Browser Automation
- Rate Limiting
- Quotas
- Concurrency
description: >-
  Scrapybara constrains usage primarily through concurrency and consumption
  quotas rather than published per-second request rates. Each plan caps the
  number of concurrent instances (Free 5, Basic 25, Pro 100) and meters monthly
  compute hours and agent credits; instances also carry a per-instance
  timeout_hours (0.01 to 24) after which they auto-terminate. Specific
  per-endpoint request-per-second limits are not published and are not
  reconciled in this artifact.
notes: >-
  Verify concurrency caps, compute-hour and agent-credit quotas, and any
  per-endpoint throttling in the Scrapybara dashboard and docs during
  reconciliation.
sources:
- https://scrapybara.com/pricing
- https://docs.scrapybara.com
- https://docs.scrapybara.com/api-reference/start
responseCodes:
  throttled: 429
limits:
- name: Concurrent Instances
  scope: account
  metric: instances
  limit: 5 (Free) / 25 (Basic) / 100 (Pro)
  notes: Maximum number of instances that can run simultaneously, set by plan tier.
- name: Compute Hours
  scope: account
  metric: hours
  limit: 10 (Free) / 100 (Basic) / 500 (Pro) per month
  notes: Monthly running-time allotment; overage billed per hour or blocked per plan.
- name: Agent Credits
  scope: account
  metric: credits
  limit: 100 (Free) / 500 (Basic) / 2500 (Pro) per month
  notes: One credit per agent step; top-ups available at $0.04 per credit on paid plans.
- name: Instance Timeout
  scope: instance
  metric: hours
  limit: 0.01 to 24
  notes: Each instance auto-terminates after its configured timeout_hours.
policies:
- name: Concurrency-Based Limiting
  description: Usage is gated primarily by concurrent-instance caps per plan rather than request-per-second throttling.
- name: Auto-Termination
  description: Instances stop automatically at timeout_hours to prevent runaway compute-hour consumption.
- 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