generated: '2026-07-19'
method: searched
source: https://docs.kyberswap.com/developer-guide/start-here/rate-limits-and-client-id
docs: https://docs.kyberswap.com/developer-guide/start-here/rate-limits-and-client-id
summary: >-
  Default rate limits are published per API and expressed in requests per
  second. Limits are tiered by the x-client-id header — requests without a
  consistent client id are subject to stricter limits. Higher tiers are granted
  by whitelisting a client id through the business development team.
limits:
- api: aggregator
  host: https://aggregator-api.kyberswap.com
  default: 3 rps
  unit: requests per second
- api: limit-order
  host: https://limit-order.kyberswap.com
  default: 10 rps
  unit: requests per second
- api: zaas
  host: https://zap-api.kyberswap.com
  default: 1 rps
  unit: requests per second
tiering:
  mechanism: x-client-id header
  value: your application or company name
  behavior: >-
    Stricter limits apply when no clientId is supplied. Use the same identifier
    consistently across both the GET route and POST route/build endpoints.
  upgrade_contact: business@kyber.network
  upgrade_process: Contact the business development team with your client identifier
    for whitelisting consideration.
exceeded:
  http_status: 429
  meaning: Too Many Requests
  remediation: Rate limit exceeded. Retry with exponential backoff.
response_headers:
  documented: false
  notes: >-
    No RateLimit-* or X-RateLimit-* response headers are documented. Clients
    should treat HTTP 429 as the only rate-limit signal.
caching_guidance:
  routes: >-
    Do not cache routes client-side for more than 5-10 seconds — quotes go stale
    as on-chain liquidity moves.
  source: https://docs.kyberswap.com/developer-guide/aggregator-api/aggregator-api-specification/evm-swaps