Langtrace integrates directly with Langchain, offering detailed, real-time insights into performance metrics such as cost, token usage, accuracy, and latency.

​
Setup

  1. Install Langtrace’s SDK and initialize the SDK in your code.

Note: You’ll need API key from Langtrace. Sign up for Langtrace if you haven’t done so already.

Python
# Install the SDK
pip install -U langtrace-python-sdk langchain langchain-chroma langchainhub
Typescript
npm install @pinecone-database/pinecone-client
  1. Setup environment variables:
Shell
export LANGTRACE_API_KEY=YOUR_LANGTRACE_API_KEY
export OPENAI_API_KEY=YOUR_OPENAI_API_KEY

​
Usage

Generate a simple output with your deployment’s model:

import os
from langtrace_python_sdk import langtrace # Must precede any llm module imports
langtrace.init(api_key = os.environ['LANGTRACE_API_KEY'])
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(model="gpt-3.5-turbo-0125")

import bs4
from langchain import hub
from langchain_chroma import Chroma
from langchain_community.document_loaders import TextLoader
from langchain_core.output_parsers import StrOutputParser
from langchain_core.runnables import RunnablePassthrough
from langchain_openai import OpenAIEmbeddings
from langchain_text_splitters import RecursiveCharacterTextSplitter

Create a vector store index and query it

# We will be loading a document from the assets folder to create embeddings and query it

loader = TextLoader("assets/soccer_rules.txt")
docs = loader.load()
text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
splits = text_splitter.split_documents(docs)
vectorstore = Chroma.from_documents(documents=splits, embedding=OpenAIEmbeddings())

# Retrieve and generate using the relevant snippets of the blog.
retriever = vectorstore.as_retriever()
prompt = hub.pull("rlm/rag-prompt")

Retrieve and generate using the relevant snippets of the document.

retriever = vectorstore.as_retriever()
prompt = hub.pull("rlm/rag-prompt")

Format our document ,setup a rag chain and query

def format_docs(docs):
    return "\n\n".join(doc.page_content for doc in docs)


rag_chain = (
    {"context": retriever | format_docs, "question": RunnablePassthrough()}
    | prompt
    | llm
    | StrOutputParser()
)

rag_chain.invoke("What is Offside??")

You can now view your traces on the Langtrace dashboard

Want to see more supported methods? Checkout the sample code in the Langtrace Langchain Python Example