Class GenerativeModel

Class for generative model APIs.

Hierarchy (View Summary)

AIModel
- GenerativeModel

Index

Constructors

constructor

new GenerativeModel(
    ai: AI,
    modelParams: ModelParams,
    requestOptions?: RequestOptions,
): GenerativeModel
Parameters
- ai: AI
- modelParams: ModelParams
- OptionalrequestOptions: RequestOptions
Returns GenerativeModel
Overrides AIModel.constructor
- Defined in models/generative-model.ts:56

Methods

countTokens

countTokens(
request: string | CountTokensRequest | (string | Part)[],
singleRequestOptions?: SingleRequestOptions,
): Promise<CountTokensResponse>
Counts the tokens in the provided request.
Parameters
- request: string | CountTokensRequest | (string | Part)[]
- OptionalsingleRequestOptions: SingleRequestOptions
Returns Promise<CountTokensResponse>
- Defined in models/generative-model.ts:150

generateContent

generateContent(
request: string | GenerateContentRequest | (string | Part)[],
singleRequestOptions?: SingleRequestOptions,
): Promise<GenerateContentResult>
Makes a single non-streaming call to the model and returns an object containing a single GenerateContentResponse.
Parameters
- request: string | GenerateContentRequest | (string | Part)[]
- OptionalsingleRequestOptions: SingleRequestOptions
Returns Promise<GenerateContentResult>
- Defined in models/generative-model.ts:70

generateContentStream

generateContentStream(
request: string | GenerateContentRequest | (string | Part)[],
singleRequestOptions?: SingleRequestOptions,
): Promise<GenerateContentStreamResult>
Makes a single streaming call to the model and returns an object containing an iterable stream that iterates over all chunks in the streaming response as well as a promise that returns the final aggregated response.
Parameters
- request: string | GenerateContentRequest | (string | Part)[]
- OptionalsingleRequestOptions: SingleRequestOptions
Returns Promise<GenerateContentStreamResult>
- Defined in models/generative-model.ts:102

startChat

startChat(startChatParams?: StartChatParams): ChatSession
Gets a new ChatSession instance which can be used for multi-turn chats.
Parameters
- OptionalstartChatParams: StartChatParams
Returns ChatSession
- Defined in models/generative-model.ts:126

`Static`normalizeModelName

normalizeModelName(modelName: string, backendType: BackendType): string
Internal
Normalizes the given model name to a fully qualified model resource name.
Parameters
- modelName: string
  The model name to normalize.
- backendType: BackendType
Returns string
The fully qualified model resource name.
Inherited from AIModel.normalizeModelName
- Defined in models/ai-model.ts:72

Properties

`Protected` `Internal`_apiSettings

_apiSettings: ApiSettings

generationConfig

generationConfig: GenerationConfig

`Readonly`model

model: string

The fully qualified model resource name (for example, gemini-3.1-flash-lite for text or publishers/google/models/gemini-3.1-flash-lite-image for image models).

`Optional`requestOptions

requestOptions?: RequestOptions

safetySettings

safetySettings: SafetySetting[]

`Optional`systemInstruction

systemInstruction?: Content

`Optional`toolConfig

toolConfig?: ToolConfig

`Optional`tools

tools?: Tool[]

Class GenerativeModel

Hierarchy (View Summary)

Index

Constructors

Methods

Properties

Constructors

constructor

Parameters

Returns GenerativeModel

Methods

countTokens

Parameters

Returns Promise<CountTokensResponse>

generateContent

Parameters

Returns Promise<GenerateContentResult>

generateContentStream

Parameters

Returns Promise<GenerateContentStreamResult>

startChat

Parameters

Returns ChatSession

StaticnormalizeModelName

Parameters

Returns string

Properties

Protected Internal_apiSettings

generationConfig

Readonlymodel

OptionalrequestOptions

safetySettings

OptionalsystemInstruction

OptionaltoolConfig

Optionaltools

Settings

On This Page

`Static`normalizeModelName

`Protected` `Internal`_apiSettings

`Readonly`model

`Optional`requestOptions

`Optional`systemInstruction

`Optional`toolConfig

`Optional`tools