LangChain Reference home pageLangChain ReferenceLangChain Reference
  • GitHub
  • Main Docs
Deep Agents
LangChain
LangGraph
Integrations
LangSmith
LangSmith
  • Client
  • Run Trees
  • Traceable
  • Evaluation
  • Schemas
  • Langchain
  • Jest
  • Vitest
  • Wrappers
  • Anonymizer
  • Jestlike
  • Vercel
  • Anthropic
  • Sandbox
⌘I

LangChain Assistant

Ask a question to get started

Enter to send•Shift+Enter new line

Menu

LangSmith
ClientRun TreesTraceableEvaluationSchemasLangchainJestVitestWrappersAnonymizerJestlikeVercelAnthropicSandbox
Language
Theme
JavaScriptlangsmithexperimentalvercelLangSmithTelemetryConfig
Interfaceā—Since v0.8

LangSmithTelemetryConfig

Copy
interface LangSmithTelemetryConfig

Properties

View source on GitHub
property
client: Client
property
extra: KVMap
property
metadata: KVMap
property
name: string
property
processChildLLMRunInputs: (inputs: KVMap) => KVMap
property
processChildLLMRunOutputs: (outputs: KVMap) => KVMap
property
processInputs: (inputs: KVMap) => KVMap
property
processOutputs: (outputs: KVMap) => KVMap
property
projectName: string
property
runType: string
property
tags: string[]
property
traceRawHttp: boolean
property
traceResponseMetadata: boolean
property
tracingEnabled: boolean

Configuration options for creating a LangSmith telemetry integration for the Vercel AI SDK.

The LangSmith client to use.

Additional RunTree configuration to pass through.

Additional metadata associated with the example.

Custom name for the root span. If not provided, defaults to the model display name or "generateText"/"streamText".

Transform inputs before logging on child LLM step spans.

Transform outputs before logging on child LLM step spans.

Transform inputs before logging on the root span.

Transform outputs before logging on the root span.

Run type for the root span. Defaults to "chain".

Tags to attach to all spans.

Whether to include raw HTTP request/response bodies.

Whether to include intermediate step details in the output.

Whether to enable tracing.