OpenInference Instrumentation for LangChain.js

This module provides automatic instrumentation for LangChain.js. which may be used in conjunction with @opentelemetry/sdk-trace-node.

Installation

npm install --save @arizeai/openinference-instrumentation-langchain

Usage

To load the Langchain instrumentation, manually instrument the @langchain/core/callbacks/manager module. The callbacks manager must be manually instrumented due to the non-traditional module structure in @langchain/core. Additional instrumentations can be registered as usual using the registerInstrumentations function.

import { NodeTracerProvider } from "@opentelemetry/sdk-trace-node";
import { LangChainInstrumentation } from "@arizeai/openinference-instrumentation-langchain";
import * as CallbackManagerModule from "@langchain/core/callbacks/manager";

const provider = new NodeTracerProvider();
provider.register();

const lcInstrumentation = new LangChainInstrumentation();
// LangChain must be manually instrumented as it doesn't have a traditional module structure
lcInstrumentation.manuallyInstrument(CallbackManagerModule);

For more information on OpenTelemetry Node.js SDK, see the OpenTelemetry Node.js SDK documentation.

Using a Custom Tracer Provider

You can specify a custom tracer provider when creating the LangChain instrumentation. This is useful when you want to use a non-global tracer provider or have more control over the tracing configuration.

import { NodeTracerProvider } from "@opentelemetry/sdk-trace-node";
import { Resource } from "@opentelemetry/resources";
import { SEMRESATTRS_PROJECT_NAME } from "@arizeai/openinference-semantic-conventions";
import { LangChainInstrumentation } from "@arizeai/openinference-instrumentation-langchain";
import * as CallbackManagerModule from "@langchain/core/callbacks/manager";

// Create a custom tracer provider
const customTracerProvider = new NodeTracerProvider({
  resource: new Resource({
    [SEMRESATTRS_PROJECT_NAME]: "my-langchain-project",
  }),
});

// Pass the custom tracer provider to the instrumentation
const lcInstrumentation = new LangChainInstrumentation({
  tracerProvider: customTracerProvider,
});

// Manually instrument the LangChain module
lcInstrumentation.manuallyInstrument(CallbackManagerModule);

Alternatively, you can set the tracer provider after creating the instrumentation:

const lcInstrumentation = new LangChainInstrumentation();
lcInstrumentation.setTracerProvider(customTracerProvider);

Compatibility

| @langchain/core Version | @arizeai/openinference-instrumentation-langchain Version | | ----------------------- | -------------------------------------------------------- | | ^1.0.0 | ^4.0.0 | | ^0.3.0 | ^4.0.0 |

This package is only tested against the 1.X versions of @langchain/core. Older versions may work but are not officially supported. For full compatibility for the 0.X versions of LangChain.js, a dedicated package called openinference-instrumentation-langchain-v0 is available.

Deprecations

LangChain v0.1 was deprecated on 2025-03-02 due to security vulnerabilities in the core package.