Source record

Published by LangChain Blog (T1.5 source). The original is at https://www.langchain.com/blog/how-to-use-rlms-in-deep-agents.

Pipeline notes

The summary and note below are generated by the signal pipeline — they are Beyond Desk’s reading, not quotations from the source.

LangChain's Deep Agents now supports Recursive Language Models (RLMs), which address context rot by letting agents write code that dispatches subagents over context chunks in a REPL rather than loading everything into one context window. Benchmarks on OOLONG show the approach holds up where turn-by-turn agents degrade.

Concrete architectural pattern for long-context agent work, with benchmark numbers. Worth reading before wiring subagent dispatch into your own setup.

Cited by

No citations on record.