LangChain's new deepagents file format cuts edit errors 15% and tokens 10%
LangChain · x · 2026-09-15
LangChain's OSS team changed how deepagents reads files, reporting that the new format reduces editfile errors by 15% and total input token usage by 10% based on their evals.
The team frames this as part of an ongoing effort on "small but mighty" agent harness optimizations, arguing these incremental improvements compound significantly in agent workflows.
More from coding & agent
- AWS Open-Sources Pizza Bot, an Inbox for Background AI Agents — CackleRooster · 2026-09-15
- Forking Running Machines: Clone a Live Doom Session into 32 Realities in One Second — bigaiguy · 2026-09-15
- Browserbase Founder: Every Online Task Will Be Agent-Assisted Within 12 Months — jeff_weinstein · 2026-09-15
- Codex Tip: Orchestrate Threads with Astra and Add 15-Minute Heartbeat Checks — reach_vb · 2026-09-15
- ElevenLabs connector ships one-click OAuth install across Claude, ChatGPT, Cursor — testingcatalog · 2026-09-15
- ElevenLabs adds Creative suite to its MCP for voice, music, video in Claude, ChatGPT, Cursor — testingcatalog · 2026-09-15