📄️ LLM
The LLM typeclass is the cornerstone of langchain-hs. It provides a unified interface for:
📄️ Document Loaders
The DocumentLoader provides a unified interface for loading Document Types from various sources.
📄️ Embeddings models
Embeddings typeclass provides interface over various Embeddings models.
📄️ Memory
In langchain-hs, Memory module provides a BaseMemory typeclass. The goal is provide types with
📄️ Output Parser
OutputParser typeclass provides an interface that parses Text (usually LLM output) into desired type.
📄️ Callbacks
Callbacks are a powerful feature in Langchain that allow you to monitor the execution of your LLMs. They provide a way to receive updates about the progress of your operations, log information on-the-fly. .
📄️ Prompt Template
Prompt templates are a powerful feature in Langchain that allow you to create dynamic prompts for your language models. They enable you to define a template with placeholders that can be filled with specific values at runtime, making it easy to generate customized prompts for different use cases.
📄️ VectorStore
VectorStore is a core component of Langchain that allows you to store and retrieve documents based on their vector representations. It is designed to work with various embedding models, enabling efficient similarity searches and retrieval of relevant documents.
📄️ Retriever
Retriever typeclass provides an even more generic interface to retrieve documents than vectorStore.
📄️ Text Splitter
Overview
📄️ Tool
Tool typeclass provides an interface to define a tool that can be used in Agents. runTool is a function that takes an input and produces an output. The Tool typeclass defines methods for executing the tool, as well as for managing the metadata associated with the tool, such as its name and description.
📄️ Agents
The core idea of agents is to use a language model to choose a sequence of actions to take. In chains, a sequence of actions is hardcoded (in code). In agents, a language model is used as a reasoning engine to determine which actions to take and in which order.
📄️ RetrievalQA
Chain for question-answering against an index.
📄️ Runnable
As of 0.0.2.0 the Runnable typeclass is pretty unstable. You can use it but expect breaking changes in the future.