Skip to main content

Class: OpenAI

OpenAI LLM implementation

Implements​

Constructors​

constructor​

• new OpenAI(init?)

Parameters​

NameType
init?Partial<OpenAI> & { azure?: AzureOpenAIConfig }

Defined in​

llm/LLM.ts:102

Properties​

additionalChatOptions​

• Optional additionalChatOptions: Omit<Partial<CompletionCreateParams>, "model" | "temperature" | "max_tokens" | "messages" | "top_p" | "streaming">

Defined in​

llm/LLM.ts:85

additionalSessionOptions​

• Optional additionalSessionOptions: Omit<Partial<ClientOptions>, "apiKey" | "timeout" | "maxRetries">

Defined in​

llm/LLM.ts:95

apiKey​

• Optional apiKey: string = undefined

Defined in​

llm/LLM.ts:91

callbackManager​

• Optional callbackManager: CallbackManager

Defined in​

llm/LLM.ts:100

maxRetries​

• maxRetries: number

Defined in​

llm/LLM.ts:92

maxTokens​

• Optional maxTokens: number

Defined in​

llm/LLM.ts:84

model​

• model: "gpt-3.5-turbo" | "gpt-3.5-turbo-16k" | "gpt-4" | "gpt-4-32k"

Defined in​

llm/LLM.ts:81

session​

• session: OpenAISession

Defined in​

llm/LLM.ts:94

temperature​

• temperature: number

Defined in​

llm/LLM.ts:82

timeout​

• Optional timeout: number

Defined in​

llm/LLM.ts:93

topP​

• topP: number

Defined in​

llm/LLM.ts:83

Methods​

chat​

▸ chat(messages, parentEvent?): Promise<ChatResponse>

Get a chat response from the LLM

Parameters​

NameType
messagesChatMessage[]
parentEvent?Event

Returns​

Promise<ChatResponse>

Implementation of​

LLM.chat

Defined in​

llm/LLM.ts:173

complete​

▸ complete(prompt, parentEvent?): Promise<ChatResponse>

Get a prompt completion from the LLM

Parameters​

NameTypeDescription
promptstringthe prompt to complete
parentEvent?Event-

Returns​

Promise<ChatResponse>

Implementation of​

LLM.complete

Defined in​

llm/LLM.ts:214

mapMessageType​

▸ mapMessageType(messageType): "function" | "user" | "assistant" | "system"

Parameters​

NameType
messageTypeMessageType

Returns​

"function" | "user" | "assistant" | "system"

Defined in​

llm/LLM.ts:156