Skip to main content
Open In ColabOpen on GitHub

Galaxia Retriever

Galaxia is GraphRAG solution, which automates document processing, knowledge base (Graph Language Model) creation and retrieval: galaxia-rag

To use Galaxia first upload your texts and create a Graph Language Model here: smabbler-cloud

After the model is built and activated, you will be able to use this integration to retrieve what you need.

The module repository is located here: github

Integration details

RetrieverSelf-hostCloud offeringPackage
Galaxia Retrieverlangchain-galaxia-retriever

Setup

Before you can retrieve anything you need to create your Graph Language Model here: smabbler-cloud

following these 3 simple steps: rag-instruction

Don't forget to activate the model after building it!

Installation

The retriever is implemented in the following package: pypi

%pip install -qU langchain-galaxia-retriever

Instantiation

from langchain_galaxia_retriever.retriever import GalaxiaRetriever

gr = GalaxiaRetriever(
api_url="beta.api.smabbler.com",
api_key="<key>", # you can find it here: https://beta.cloud.smabbler.com/user/account
knowledge_base_id="<knowledge_base_id>", # you can find it in https://beta.cloud.smabbler.com , in the model table
n_retries=10,
wait_time=5,
)

Usage

result = gr.invoke("<test question>")
print(result)

Use within a chain

from langchain_core.output_parsers import StrOutputParser
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.runnables import RunnablePassthrough

prompt = ChatPromptTemplate.from_template(
"""Answer the question based only on the context provided.

Context: {context}

Question: {question}"""
)


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


chain = (
{"context": gr | format_docs, "question": RunnablePassthrough()}
| prompt
| llm
| StrOutputParser()
)
chain.invoke("<test question>")

API reference

For more information about Galaxia Retriever check its implementation on github github


Was this page helpful?