Python APIlangchain

Core

Instrumentation

Python · Instrumentation

openinference-instrumentation-langchain

OpenInference LangChain Instrumentation

Module
openinference.instrumentation.langchain
Version
0.1.87
Python
>=3.10, <3.15
Instruments
langchain_core >= 0.3.9

Install

pip install openinference-instrumentation-langchain

Usage

Instrument once, at startup, before the library is used. The LangChainInstrumentor reference below lists the options.

from opentelemetry.sdk.trace import TracerProvider
from openinference.instrumentation import TraceConfig
from openinference.instrumentation.langchain import LangChainInstrumentor

tracer_provider = TracerProvider()
# ...add a span processor that exports to Phoenix or any OTLP backend.

LangChainInstrumentor().instrument(
    tracer_provider=tracer_provider,
    config=TraceConfig(hide_inputs=True),  # optional
)

openinference.instrumentation.langchain

LangChainInstrumentor

classSource ↗
class LangChainInstrumentor(BaseInstrumentor)

An instrumentor for LangChain

instrument

method
def instrument(
    *,
    tracer_provider: Optional[TracerProvider] = None,
    config: Optional[TraceConfig] = None,
    skip_dep_check: bool = False,
) -> None

Start tracing langchain_core: patch it so its calls make OpenInference spans. Calling it again does nothing. Inherited from OpenTelemetry’s BaseInstrumentor.

Parameters

NameDescription
tracer_provider

The tracer provider the spans are made with. Defaults to the global one, opentelemetry.trace.get_tracer_provider().

config

What the spans record: hide inputs, outputs or messages, or limit image sizes. Defaults to TraceConfig(), which reads the OPENINFERENCE_* environment variables. See TraceConfig.

skip_dep_check

Instrument even if the installed version of langchain_core is outside the supported range.

uninstrument

method
def uninstrument() -> None

Stop tracing: undo the patches instrument() made. Inherited from OpenTelemetry’s BaseInstrumentor.

instrumentation_dependencies

methodSource ↗
def instrumentation_dependencies() -> Collection[str]

get_span

methodSource ↗
def get_span(run_id: UUID) -> Optional[Span]

get_ancestors

methodSource ↗
def get_ancestors(run_id: UUID) -> List[Span]

get_current_span

functionSource ↗
def get_current_span() -> Optional[Span]

get_ancestor_spans

functionSource ↗
def get_ancestor_spans() -> List[Span]

Retrieve the ancestor spans for the current LangChain run.

This function traverses the LangChain run tree from the current run’s parent up to the root, collecting the spans associated with each ancestor run. The list is ordered from the immediate parent of the current run to the root of the run tree.