Skip to main content

Rate Limits

Memory API request frequency limits

Memory API rate limits are at the Alibaba Cloud account level.
APILimit
All endpointsUp to 3000 QPM total
Factual memory add120 QPM
Factual memory search300 QPM
When exceeded, HTTP 429 is returned. Use exponential backoff for retries. To request a higher rate limit, submit a ticket.
See Billing for pricing details.
Agent Development
Agent Context
Agent Governance & Evolution
Store
Application Marketplace