Connect to Google's generative AI embeddings service using theGoogleGenerativeAIEmbeddings class, found in thelangchain-google-genai package.

This will help you get started with Google's Generative AI embedding models (like Gemini) using LangChain. For detailed documentation onGoogleGenerativeAIEmbeddings features and configuration options, please refer to theAPI reference.

Overview​

Integration details​

Setup​

To access Google Generative AI embedding models you'll need to create a Google Cloud project, enable the Generative Language API, get an API key, and install thelangchain-google-genai integration package.

Credentials​

To use Google Generative AI models, you must have an API key. You can create one in Google AI Studio. See theGoogle documentation for instructions.

Once you have a key, set it as an environment variableGOOGLE_API_KEY:

import getpass
import os

ifnot os.getenv("GOOGLE_API_KEY"):
    os.environ["GOOGLE_API_KEY"]= getpass.getpass("Enter your Google API key: ")

To enable automated tracing of your model calls, set yourLangSmith API key:

# os.environ["LANGSMITH_TRACING"] = "true"
# os.environ["LANGSMITH_API_KEY"] = getpass.getpass("Enter your LangSmith API key: ")

Installation​

%pip install--upgrade--quiet  langchain-google-genai

Usage​

from langchain_google_genaiimport GoogleGenerativeAIEmbeddings

embeddings= GoogleGenerativeAIEmbeddings(model="models/gemini-embedding-exp-03-07")
vector= embeddings.embed_query("hello, world!")
vector[:5]

[-0.024917153641581535,
 0.012005362659692764,
 -0.003886754624545574,
 -0.05774897709488869,
 0.0020742062479257584]

Batch​

You can also embed multiple strings at once for a processing speedup:

vectors= embeddings.embed_documents(
[
"Today is Monday",
"Today is Tuesday",
"Today is April Fools day",
]
)
len(vectors),len(vectors[0])

(3, 3072)

Indexing and Retrieval​

Embedding models are often used in retrieval-augmented generation (RAG) flows, both as part of indexing data as well as later retrieving it. For more detailed instructions, please see ourRAG tutorials.

Below, see how to index and retrieve data using theembeddings object we initialized above. In this example, we will index and retrieve a sample document in theInMemoryVectorStore.

# Create a vector store with a sample text
from langchain_core.vectorstoresimport InMemoryVectorStore

text="LangChain is the framework for building context-aware reasoning applications"

vectorstore= InMemoryVectorStore.from_texts(
[text],
    embedding=embeddings,
)

# Use the vectorstore as a retriever
retriever= vectorstore.as_retriever()

# Retrieve the most similar text
retrieved_documents= retriever.invoke("What is LangChain?")

# show the retrieved document's content
retrieved_documents[0].page_content

'LangChain is the framework for building context-aware reasoning applications'

Task type​

GoogleGenerativeAIEmbeddings optionally support atask_type, which currently must be one of:

• SEMANTIC_SIMILARITY: Used to generate embeddings that are optimized to assess text similarity.
• CLASSIFICATION: Used to generate embeddings that are optimized to classify texts according to preset labels.
• CLUSTERING: Used to generate embeddings that are optimized to cluster texts based on their similarities.
• RETRIEVAL_DOCUMENT,RETRIEVAL_QUERY,QUESTION_ANSWERING, andFACT_VERIFICATION: Used to generate embeddings that are optimized for document search or information retrieval.
• CODE_RETRIEVAL_QUERY: Used to retrieve a code block based on a natural language query, such as sort an array or reverse a linked list. Embeddings of the code blocks are computed usingRETRIEVAL_DOCUMENT.

By default, we useRETRIEVAL_DOCUMENT in theembed_documents method andRETRIEVAL_QUERY in theembed_query method. If you provide a task type, we will use that for all methods.

%pip install--upgrade--quiet  matplotlib scikit-learn

from langchain_google_genaiimport GoogleGenerativeAIEmbeddings
from sklearn.metrics.pairwiseimport cosine_similarity

query_embeddings= GoogleGenerativeAIEmbeddings(
    model="models/gemini-embedding-exp-03-07", task_type="RETRIEVAL_QUERY"
)
doc_embeddings= GoogleGenerativeAIEmbeddings(
    model="models/gemini-embedding-exp-03-07", task_type="RETRIEVAL_DOCUMENT"
)

q_embed= query_embeddings.embed_query("What is the capital of France?")
d_embed= doc_embeddings.embed_documents(
["The capital of France is Paris.","Philipp is likes to eat pizza."]
)

for i, dinenumerate(d_embed):
print(f"Document{i+1}:")
print(f"Cosine similarity with query:{cosine_similarity([q_embed],[d])[0][0]}")
print("---")

Document 1
Cosine similarity with query: 0.7892893360164779
---
Document 2
Cosine similarity with query: 0.5438283285204146
---

API Reference​

For detailed documentation onGoogleGenerativeAIEmbeddings features and configuration options, please refer to theAPI reference.

Additional Configuration​

You can pass the following parameters to ChatGoogleGenerativeAI in order to customize the SDK's behavior:

• client_options:Client Options to pass to the Google API Client, such as a customclient_options["api_endpoint"]
• transport: The transport method to use, such asrest,grpc, orgrpc_asyncio.

Related​

• Embedding modelconceptual guide
• Embedding modelhow-to guides