LangChain ships LangSmith Tuned Evaluators, starting with Perceived Error
LangSmith Tuned Evaluators are managed, versioned evaluators that auto-attach quality feedback to production traces and threads. The launch evaluators — Perceived Error — flag threads where the agent made a mistake, misunderstood a request, or derailed; the specialized post-trained judge reportedly outperformed every frontier model in LangChain's benchmark at up to 98% lower cost in some workloads. Available now for Plus and Cloud Enterprise plans in the US.
Signal context
LANGCHAIN.COMLangChain · Chronicle
7Same-week signals
± 7 daysNearby on the map
Same continentField notes
FAQWhat did LangChain announce on AUG 18, 2026?
LangSmith Tuned Evaluators are managed, versioned evaluators that auto-attach quality feedback to production traces and threads. The launch evaluators — Perceived Error — flag threads where the agent made a mistake, misunderstood a request, or derailed; the specialized post-trained judge reportedly outperformed every frontier model in LangChain's benchmark at up to 98% lower cost in some workloads. Available now for Plus and Cloud Enterprise plans in the US.
What is LangChain?
LangChain the company ships an agent engineering platform on top of the open-source LangChain and LangGraph frameworks. LangSmith is the commercial observability and evaluation layer — tracing, automatic evaluators (Perceived Error and more), preview deployments from PR branches, and AgentCore Payments middleware that lets agents transact through the x402 flow with session-level spending limits.
Where is the official source for this announcement?
It was published by LangChain on AUG 18, 2026 via langchain.com, the company's official channel. AgentMaps cites the primary source for every charted signal.