langchain-google-genai

Description

`langchain-google-genai`

LangChain integration for Google's Generative AI models, providing access to Gemini models via both the Gemini Developer API and Vertex AI.

Vertex AI consolidation

As of langchain-google-genai 4.0.0, this package uses the consolidated google-genai SDK instead of the legacy google-ai-generativelanguage SDK.

This migration brings support for Gemini models both via the Gemini Developer API and Gemini API in Vertex AI, superseding certain classes in langchain-google-vertexai, such as ChatVertexAI. Refer to the provider docs and release notes for more information.

Modules

Usage documentation Refer to the docs for a high-level guide on how to use each module. These reference pages contain auto-generated API documentation for each module, focusing on the "what" rather than the "how" or "why" (i.e. no end-to-end tutorials or conceptual overviews).

ChatGoogleGenerativeAI

Gemini chat models (primary interface).

GoogleGenerativeAI

(Legacy) Google text completion abstraction.

GoogleGenerativeAIEmbeddings

Gemini embedding models.

create_context_cache

Utility to create a context cache for reusing large content across requests.

Classes

Class

GoogleGenerativeAIEmbeddings

Google Generative AI Embeddings.

Text-only

While gemini-embedding-2-preview natively supports multimodal inputs (text, images, video, audio, and PDFs) via the Google GenAI SD

Class

GoogleGenerativeAIError

Custom exception class for errors associated with the Google GenAI API.

Class

GoogleGenerativeAI

Google GenerativeAI text completion large language models (legacy LLMs).

from langchain_google_genai import GoogleGenerativeAI

llm = GoogleGenerative

<!--/ADMON-->

Class

ChatGoogleGenerativeAIError

Wrapper exception class for errors associated with the Google GenAI API.

Raised when there are specific issues related to the Google GenAI API usage in the ChatGoogleGenerativeAI class, such as u

Class

GoogleContextOverflowError

ClientError raised when input exceeds Google's context limit.

Class

GoogleAuthenticationError

Google authentication error classified as a LangChain model error.

Class

GooglePermissionDeniedError

Google permission error classified as a LangChain model error.

Class

GoogleInvalidRequestError

Google invalid-request error classified as a LangChain model error.

Class

GoogleModelNotFoundError

Google not-found error classified as a LangChain model error.

Class

GoogleRateLimitError

Google rate-limit error classified as a LangChain model error.

Class

GoogleAPIError

Google server error classified as a LangChain model error.

Class

ChatGoogleGenerativeAI

Google GenAI chat model integration.

Class

Route

Image loading route.

Determines how media content is loaded and processed based on the input string format.

Class

ImageBytesLoader

Loads media bytes from multiple sources given a string.

(Despite the class name, this loader supports multiple media types including images, PDFs, audio, and video files.)

Currently supported:

Functions

Function

convert_to_genai_function_declarations

Convert tools to google-genai Tool objects.

Each special tool type (google_search, google_maps, url_context, etc.) must be in its own Tool object due to protobuf oneof constraints.

Function

create_context_cache

Creates a context cache for the specified model and content.

Context caching allows you to store and reuse content (e.g., PDFs, images) for faster processing. This is useful when you have large amoun

Function

get_user_agent

Returns a custom user agent header.

Function

image_bytes_to_b64_string

Encodes image bytes into a b64 encoded string.

Function

translate_citations_to_grounding_metadata

Translate LangChain Citations to Google AI grounding metadata format.

Modules

Module