All Classes and Interfaces
Class
Description
Class responsible for calling the language model and deciding the action.
Agent's action to take.
Consists of an agent using tools.
Agent's return value.
Parse text into agent action/finish.
Type of message that is spoken by the AI.
Chain that makes API calls and summarizes the responses to answer a question.
Information about a data source attribute.
Base interface for chat message history, See `ChatMessageHistory` for default implementation.
Base interface for chains combining documents.
Base interface for transforming documents.
BaseLanguageModel is an interface for interacting with a language model.
LLM wrapper should take in a prompt and return a string.
Interface for loading documents.
Base interface for memory in chains.
Message object.
Wrapper around OpenAI large language models.
Class to parse the output of an LLM call.
Base class for all prompt templates, returning a prompt.
Base interface for retrievers.
Base Agent class.
Interface LangChain tools must implement.
Base Toolkit representing a collection of related tools.
Callbacks
Base interface that all chains should implement.
Implementation of splitting text that looks at characters.
Output of a single generation.
ChatGLM LLM service.
Type of message with arbitrary speaker.
interface for database supported chat message repository;
Wrapper around OpenAI Chat large language models.
Class that contains all relevant information for a Chat Result.
Parse out comma separated lists.
A comparison to a value.
Prompt collection that goes through conditionals.
Buffer for storing conversation memory.
Chain to have a conversation and load context from memory.
database based chat message history;
Load documents from a directory.
Interface for interacting with a document.
Interface for embedding models.
Prompt template that contains few shot examples.
A filtering expression.
Output of a single generation.
A Generation chunk, which can be concatenated with other Generation chunks.
Wrapper for Google Search API.
Type of message that is spoken by the human.
Load agent.
Tool that is run when invalid tool name is encountered by agent.
LangChainException
Line type as typed dict.
Class to parse the output of an LLM call to a list.
Base LLM abstract class.
Chain to run queries against LLMs
Chain that interprets a prompt and executes python code to do math.
Class that contains all relevant information for an LLM Result.
Load tools.
Implementation of splitting markdown files based on specified headers.
Math utils.
Since the token we input to LLM is limited, the history messages we add into gpt request is also limited.
Prompt template that assumes variable is already list of messages.
Initialize wrapper around the milvus vector database.
Output parser that just returns the text as is.
Loader that loads Notion directory dump.
Ollama locally run large language models.
OpenAI
Wrapper around OpenAI Chat large language models.
Wrapper around OpenAI embedding models.
A logical operation over other directives.
Enumerator of the operations.
Exception that output parsers should raise to signify a parsing error.
Logic for converting internal query language elements to valid filters.
Prompt
Schema to represent a prompt for an LLM.
PromptValue
Implementation of splitting text that looks at characters.
a simple wrapper for DataBaseChatMessageHistory with redisChatMessageRepository;
Wrapper around requests to handle auth and async.
Chain for question-answering against an index.
Retriever that wraps around a vector store and uses an LLM to generate the vector store queries.
Wrapper around SerpAPI.
SQLAlchemy wrapper around a database.
Chain for interacting with SQL Database.
Chain for querying SQL database that is a sequential chain.
String prompt should expose the format method, returning a prompt.
StringPromptValue
Chain that combines documents by stuffing into context.
Type of message that is a system message.
Load text files.
Lightweight wrapper around requests library.
Interface for splitting text into chunks.
Tool that takes in function or coroutine directly.
Utility functions for working with vectors and vectorStores.
Defines interface for IR translation using visitor pattern.
Agent for the MRKL chain.