Alibaba Cloud Bailian Memory — a managed service that gives agents cross-session long-term memory, with APIs to write, search, and manage memories.
LLMs are limited by context windows and cannot retain information across sessions. Memory automatically extracts key information from conversations and persists it, then retrieves relevant memories via semantic search and injects them into context in subsequent conversations.
Memory is exposed as an API, so any application can integrate it and multiple applications can share the same memory. Every account comes with a default memory that works without any setup.
Memory started commercial billing on August 20, 2026, 10:00 AM (Beijing Time), with a free quota included. See Billing.
Quickstart
Write, view, and search memories in 3 steps
Core Concepts
Fragments, profiles, and how it works
Create Memory
Use the default memory, or create and delete custom ones
Configure Rules
Set extraction rules for fragments and user profiles
User Profiles
Create profile schemas and extract attributes via API
Manage Memories
View, search, update, and delete memories
Billing
Pro/Lite pricing, free quota, and examples
API Reference
Parameters for all endpoints