Skip to content
masterclassBuddyTech Buzz5 fresh blogs · 5:00 AM IST

API reference

LangSmith API reference

observability, evaluation, tracing, and prompt engineering platform for production LLM applications

Identifiers
48
With full guides
0
Types
8

Topics covered

  • Client
  • traceable decorator
  • wrap_openai
  • RunTree spans
  • Evaluation datasets
  • expect assertions
  • Online evaluators
  • Pytest plugin

Class

NameWhat it does
ClientClass
AsyncClientClass
RunTreeClass
ApiKeyAuthClass
DynamicRunEvaluatorClass
ExactMatchEvaluatorClass
CriteriaEvaluatorClass
EmbeddingDistanceEvaluatorClass

Function

NameWhat it does
wrap_openaiFunction
evaluateFunction
expectFunction
tracing_contextFunction
as_run_evaluatorFunction
aevaluateFunction
get_current_run_treeFunction
create_presigned_urlFunction
run_on_datasetFunction
wrap_anthropicFunction
create_llm_as_judgeFunction
create_datasetFunction
create_feedbackFunction
push_promptFunction
pull_promptFunction

Interface

NameWhat it does
RunEvaluatorInterface
StringEvaluatorInterface

Type Alias

NameWhat it does
EvaluationResultType Alias
RunType Alias
DatasetType Alias
FeedbackType Alias
ExampleType Alias
TraceType Alias
FeedbackSourceType Alias
FeedbackCategoryType Alias
ScoreTypeType Alias

Const

NameWhat it does
LANGSMITH_TRACINGConst
LANGSMITH_API_KEYConst
LANGSMITH_PROJECTConst
LANGSMITH_ENDPOINTConst

Enum

NameWhat it does
RunTypeEnum

Decorator

NameWhat it does
traceableDecorator
unitDecorator
pytest.mark.langsmithDecorator

Module

NameWhat it does
pytest-pluginModule
Metadata & tagsModule
ThreadsModule
Distributed tracingModule
Pairwise evaluationModule
Vitest & JestModule

Search across every package and filter by type in your dashboard.

Start learning