Interface AutoChain tools must implement.

Wrapper around OpenAI Chat large language models. To use, you should have the ``openai`` python package installed, and the environment variab

Standardized data class for each test case for BastTest

Type of message that is spoken by the human.

Agent's return value.

Buffer for storing conversation memory and an in-memory kv store.

Buffer for storing conversation memory and an in-memory kv store.

Type of message that is spoken by the AI.

Agent's action to take.

Default chain with take_next_step implemented It handles a few common error cases with agent, such as taking repeated action with same in

Use ChromaDB as internal search tool

Use LanceDB as the internal search tool LanceDB is a vector database that supports vector search. Args: uri: the uri of the dat

Use Pinecone as the internal search tool

Type of message that is a system message.

Type of message that is a function message.

Output of a single generation.

Format prompt with string Template and dictionary of variables

Class that contains all relevant information for an LLM Result.

Store conversation info in redis memory.

Exception when tool fails to run

Conversational agent who can use tools available to make a conversation by following the conversational planning prompt

Wrapper for Google Search API. Adapted from: Instructions adapted from https://stackoverflow.com/questions/ 37083058/ programmatically-se

Wrapper chain instantiate from LangChain's Chain object to match AutoChain interface

Text encoder using OpenAI Model

Base interface that all chains should implement. Chain will standardize inputs and outputs, the main entry pointy is the run function.

Wrapper around OpenAI Chat large language models. To use, you should have the ``openai`` python package installed, and the environment variab

Base interface for memory in chains.

Message object.

Configuration for this pydantic object.

Configuration for this pydantic object.

Configuration for this pydantic object.

Configuration for this pydantic object.

Configuration for this pydantic object.

Configuration for this pydantic object.

Configuration for this pydantic object.

Configuration for this pydantic object.

Configuration for this pydantic object.

Tool that has capability to query the Google Search API and get back json.

Huggingface model that supports text-generation task Example: .. code-block:: python from autochain.models.huggingface_text_generati

Agent supports function calling natively in OpenAI, which leverage function message to determine which tool should be used When tool is n

Exception that output parsers should raise to signify a parsing error. This exists to differentiate parsing errors from other code or execution e

Generate a terminal widget.