Theia API Documentation v1.77.0
    Preparing search index...

    A model served by the ChatGPT subscription of the signed in user. The endpoint only implements the streaming Response API, so unlike the plain OpenAI model there is no Chat Completions path and no fallback to one.

    Implements

    Index
    • Parameters

      • id: string

        the unique id for this language model. It will be used to identify the model in the UI.

      • model: string

        the model id as it is used by the ChatGPT endpoint

      • status: LanguageModelStatus
      • credentials: () => Promise<ChatGptCredentials | undefined>

        returns the credentials of the signed in user, refreshing the access token when needed

      • responseApiUtils: ChatGptResponseApiUtils
      • modelUtils: OpenAiModelUtils
      • maxRetries: number = 3

        the maximum number of retry attempts when a request fails

      • Optionalproxy: string
      • OptionalreasoningSupport: ReasoningSupport
      • OptionalmaxInputTokens: number
      • serverSideCompactionEnabledByDefault: boolean = true
      • OptionalserverSideCompactionTokenThresholdByDefault: number

      Returns ChatGptModel

    credentials: () => Promise<ChatGptCredentials | undefined>

    returns the credentials of the signed in user, refreshing the access token when needed

    id: string

    the unique id for this language model. It will be used to identify the model in the UI.

    maxInputTokens?: number
    maxRetries: number = 3

    the maximum number of retry attempts when a request fails

    model: string

    the model id as it is used by the ChatGPT endpoint

    modelUtils: OpenAiModelUtils
    proxy?: string
    reasoningSupport?: ReasoningSupport
    responseApiUtils: ChatGptResponseApiUtils
    serverSideCompactionEnabledByDefault: boolean = true
    serverSideCompactionSupport: true

    Whether this model supports provider-native server-side compaction (capability, distinct from whether it is activated).

    serverSideCompactionTokenThresholdByDefault?: number
    serverTools: ServerToolDescriptor[] = OPENAI_SERVER_TOOLS

    The endpoint serves Codex, which offers the same server-side web search, so it is offered for every model rather than derived from the endpoint the way the plain OpenAI model does.

    vendor: "chatgpt" = 'chatgpt'