Model Releases
it's always funny to me when these bot posts say we *just* launched deepagents... like nah man we've been carving this stone for close to a …
it's always funny to me when these bot posts say we *just* launched deepagents... like nah man we've been carving this stone for close to a year now! Long-running agents almost always fail because the
it's always funny to me when these bot posts say we just launched deepagents... like nah man we've been carving this stone for close to a year now! Long-running agents almost always fail because they suffocate in their own context window. LangChain just open-sourced Deep Agents, a new framework built explicitly to solve state and memory for complex tasks like research and coding. It shifts context management from manual scaf…
Source: Harrison Chase (X) | 2026-05-03