Async Await Support For Concurrent Llm Operations

via “asynchronous memory operations with async/await support”

Universal memory layer for AI Agents

Unique: Provides full async/await support (AsyncMemory, AsyncMemoryClient) with non-blocking I/O for all operations (LLM calls, vector store queries, graph operations), enabling seamless integration with async frameworks without thread pools or blocking calls.

vs others: More efficient than sync-based memory systems in async contexts because it avoids thread pool overhead and enables true concurrent execution, and more practical than manual async wrappers because async is built into the core API.