LangChain introduces asynchronous sub-agents for Deep Agents
The story so far
LangChain introduces asynchronous sub-agents in Deep Agents: a resident orchestrating agent handles user conversations and manages context, while sub-agents act as background work units to execute long tasks in parallel and report results upon completion. LangChain states that this allows the agent to handle increasing workloads as the number of background work units grows, rather than depending on the length of a single cycle. This feature is available in the Python and JavaScript versions of Deep Agents and Managed Deep Agents; in Managed Deep Agents, setting sub-agents to asynchronous requires only one field. Task notifications are still in beta, and the API may change. LangChain indicates that the async-deep-agents repository provides a runnable supervisor, including researcher and coder sub-agents, offering Python and TypeScript versions.
AI-written from the reports · Updated 12 days ago
Coverage timeline
Follow the coverage to see every side of the story.
- LangChain BlogFeaturedAgents Are Going Async
LangChain introduces asynchronous sub-agents in Deep Agents, where a resident orchestrating agent interfaces with users, and the sub-agents execute long tasks in the background in parallel and report results upon completion.