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: Exa
providerId: exa-ai
created: '2026-05-25'
modified: '2026-05-25'
reconciled: true
tags:
  - AI
  - Search
  - Rate Limiting
  - Quotas
description: Reconciled rate limits for the Exa Search, Contents, Answer, Research, Monitors, Agent, and Websets APIs. Per-team rate limits enforced via x-api-key, with tunable per-key limits through the Team Management API.
sources:
  - https://exa.ai/docs/reference/search-api-guide
  - https://exa.ai/docs/api-reference/openapi
  - https://exa.ai/pricing
headers:
  limit: x-ratelimit-limit
  remaining: x-ratelimit-remaining
  reset: x-ratelimit-reset
  retryAfter: retry-after
responseCodes:
  throttled: 429
  quotaExceeded: 402
algorithm: token-bucket
notes:
  - Exa publishes per-team rate limits, not per-user. Teams can subdivide limits across API keys via the Team Management API (POST /api-keys with `rateLimit` and `budget`).
  - The free tier provides 1,000 requests/month with the same per-second limits as paid plans; throttling kicks in above sustained burst rates.
  - Deep Search and Agent endpoints have higher per-request costs and longer-running operations, so concurrency is the practical limit rather than RPM.
limits:
  - tier: Free
    surface: All public endpoints
    monthlyRequests: 1000
    rpm: null
    description: 1,000 free requests/month shared across Search, Contents, Answer, Research, Monitors, and Agent.
  - tier: Search (paid)
    surface: /search, /contents, /answer
    monthlyRequests: -1
    rpm: null
    description: Standard token-bucket throughput; per-team rate limit configurable, default suitable for production agent workloads.
  - tier: Deep Search / Research
    surface: /research/v1
    monthlyRequests: -1
    concurrency: null
    description: Long-running asynchronous research tasks; concurrency limit governs how many tasks can run in parallel per team.
  - tier: Monitors
    surface: /monitors, /v0/monitors
    monthlyRequests: -1
    description: Scheduled monitors run on internal cadence; webhook delivery is throttled per webhook endpoint.
  - tier: Agent
    surface: /agent/runs
    monthlyRequests: -1
    description: Effort-mode-priced agent runs; concurrency limit governs parallel runs per team.
  - tier: Websets
    surface: /v0/websets
    monthlyRequests: -1
    description: Asynchronous webset creation, enrichment, and search; long-running tasks tracked via Events API.
perKeyControls:
  description: Teams can mint per-key rate limits and per-key monthly budgets via the Team Management API (POST /api-keys).
  endpoint: https://admin-api.exa.ai/team-management/api-keys
  fields:
    - rateLimit
    - budget
quotas:
  - name: free-monthly-requests
    metric: request
    limit: 1000
    timeFrame: month
    appliesTo: Free tier teams
billing:
  paymentMethods:
    - credit-card
    - invoice (Enterprise)
  spendLimitBehavior: API key budget enforces hard stop; team-level overage uses configured payment method.