langchain-opea

This package contains the LangChain integrations for OPENAI Compatible OPEA Microservices.

Installation

You can install LangChain OPEA package in several ways:

Install from Source

To install the package from the source, run:

pip install poetry && poetry install --with test

Install from Wheel Package

To install the package from a pre-built wheel, run:

1. Build the Wheels: Ensure the wheels are built using Poetry.

   poetry build

2. Install via Wheel File: Install the package using the generated wheel file.

   pip install dist/langchain_opea-0.1.0-py3-none-any.whl

Additionally, you’ll need to run the OpenAI compatible model servers locally. Please refer to these instructions for guidance.

Chat Models

ChatOPEA class exposes OPENAI Compatible chat models from OPEA.

from langchain_opea import ChatOPEA

llm = ChatOPEA(
    model="Intel/neural-chat-7b-v3-3", opea_api_key="my_secret_value", opea_api_base="http://localhost:9009/v1"
)
llm.invoke("Sing a ballad of LangChain.")

Embeddings

OPEAEmbeddings class exposes OPENAI Compatible embeddings from OPEA.

from langchain_opea import OPEAEmbeddings

embeddings = OPEAEmbeddings(
    model="BAAI/bge-large-en-v1.5",
    opea_api_key="my_secret_value",
    opea_api_base="http://localhost:6006/v1",
)
embeddings.embed_query("What is the meaning of life?")

LLMs

OPEALLM class exposes OPENAI Compatible LLMs from OPEA.

from langchain_opea import OPEALLM

llm = OPEALLM(
    model="Intel/neural-chat-7b-v3-3", opea_api_key="my_secret_value", opea_api_base="http://localhost:9009/v1"
)
llm.invoke("The meaning of life is")

Check out Samples for more examples using the OPEA Langchain package.