Input to a Google Vertex AI LLM class.

interface VertexAIInput {
    apiKey?: string;
    apiVersion?: string;
    authOptions?: GoogleAuthOptions<JSONClient>;
    cache?: boolean | BaseCache<Generation[]>;
    callbackManager?: CallbackManager;
    callbacks?: Callbacks;
    concurrency?: number;
    convertSystemMessageToHumanContent?: boolean;
    endpoint?: string;
    location?: string;
    maxConcurrency?: number;
    maxOutputTokens?: number;
    maxRetries?: number;
    mediaManager?: MediaManager;
    metadata?: Record<string, unknown>;
    model?: string;
    modelName?: string;
    onFailedAttempt?: FailedAttemptHandler;
    platformType?: GooglePlatformType;
    responseMimeType?: GoogleAIResponseMimeType;
    safetyHandler?: GoogleAISafetyHandler;
    safetySettings?: GoogleAISafetySetting[];
    stopSequences?: string[];
    streaming?: boolean;
    tags?: string[];
    temperature?: number;
    topK?: number;
    topP?: number;
    verbose?: boolean;
}

Hierarchy

  • GoogleLLMInput
    • VertexAIInput

Properties

apiKey?: string

Some APIs allow an API key instead

apiVersion?: string

The version of the API functions. Part of the path.

authOptions?: GoogleAuthOptions<JSONClient>
cache?: boolean | BaseCache<Generation[]>
callbackManager?: CallbackManager

Use callbacks instead

callbacks?: Callbacks
concurrency?: number

Use maxConcurrency instead

convertSystemMessageToHumanContent?: boolean
endpoint?: string

Hostname for the API call (if this is running on GCP)

location?: string

Region where the LLM is stored (if this is running on GCP)

maxConcurrency?: number

The maximum number of concurrent calls that can be made. Defaults to Infinity, which means no limit.

maxOutputTokens?: number

Maximum number of tokens to generate in the completion.

maxRetries?: number

The maximum number of retries that can be made for a single call, with an exponential backoff between each attempt. Defaults to 6.

mediaManager?: MediaManager
metadata?: Record<string, unknown>
model?: string

Model to use

modelName?: string

Model to use Alias for model

onFailedAttempt?: FailedAttemptHandler

Custom handler to handle failed attempts. Takes the originally thrown error object as input, and should itself throw an error if the input error is not retryable.

platformType?: GooglePlatformType

What platform to run the service on. If not specified, the class should determine this from other means. Either way, the platform actually used will be in the "platform" getter.

responseMimeType?: GoogleAIResponseMimeType

Available for gemini-1.5-pro. The output format of the generated candidate text. Supported MIME types:

  • text/plain: Text output.
  • application/json: JSON response in the candidates.
"text/plain"
safetyHandler?: GoogleAISafetyHandler
safetySettings?: GoogleAISafetySetting[]
stopSequences?: string[]
streaming?: boolean

Whether or not to stream.

false
tags?: string[]
temperature?: number

Sampling temperature to use

topK?: number

Top-k changes how the model selects tokens for output.

A top-k of 1 means the selected token is the most probable among all tokens in the model’s vocabulary (also called greedy decoding), while a top-k of 3 means that the next token is selected from among the 3 most probable tokens (using temperature).

topP?: number

Top-p changes how the model selects tokens for output.

Tokens are selected from most probable to least until the sum of their probabilities equals the top-p value.

For example, if tokens A, B, and C have a probability of .3, .2, and .1 and the top-p value is .5, then the model will select either A or B as the next token (using temperature).

verbose?: boolean