googleai_dart library

Unofficial Dart client for the Google AI Gemini Developer API and Vertex AI Gemini API with unified interface.

Classes

AgentConfig: Base class for agent configurations.
AllowedTools: Configuration for allowed tools.
Annotation: Citation information for model-generated content.
ApiKeyCredentials: API key authentication credentials.
ApiKeyProvider: Simple API key provider with static credentials.
AttributionSourceId: Identifier for the source contributing to this attribution.
AudioContent: An audio content block.
AudioDelta: An audio delta update.
AudioTranscriptionConfig: Configuration for audio transcription.
AuthCredentials: Authentication credentials returned by AuthProvider.
AuthProvider: Authentication provider interface for the GoogleAI client.
AuthToken: Ephemeral authentication token for secure client-side API access.
AutomaticActivityDetection: Configuration for automatic voice activity detection (VAD).
BatchEmbedContentsRequest: Batch request to get embeddings from the model for a list of prompts.
BatchEmbedContentsResponse: The response to a BatchEmbedContentsRequest.
BatchStats: Stats about a batch operation.
BearerTokenCredentials: Bearer token authentication credentials.
BearerTokenProvider: Simple bearer token provider with static credentials.
BidiGenerateContentClientContent: Client content message for sending conversation context.
BidiGenerateContentClientMessage: Base class for client-to-server WebSocket messages.
BidiGenerateContentRealtimeInput: Real-time input message for streaming audio, video, or text.
BidiGenerateContentServerContent: Server content message containing model output.
BidiGenerateContentServerMessage: Base class for server-to-client WebSocket messages.
BidiGenerateContentSetup: Initial session setup message.
BidiGenerateContentSetupComplete: Session setup complete message.
BidiGenerateContentToolCall: Tool call request message.
BidiGenerateContentToolCallCancellation: Tool call cancellation message.
BidiGenerateContentToolResponse: Tool response message.
Blob: Inline binary data (e.g., image, audio) encoded as base64.
CachedContent: Cached content for reuse across requests.
CachedContentUsageMetadata: Usage metadata for cached content.
Candidate: A generated response from the model.
ChunkingConfig: Parameters for telling the service how to chunk the file.
CitationMetadata: Citation sources for generated content.
CitationSource: A citation source.
CodeExecutionCallContent: A code execution call content block.
CodeExecutionCallDelta: A code execution call delta update.
CodeExecutionResult: Result of executing the ExecutableCode.
CodeExecutionResultContent: A code execution result content block.
CodeExecutionResultDelta: A code execution result delta update.
CodeExecutionResultPart: Result from code execution.
CodeExecutionTool: A tool that allows the model to execute code.
ComputerUse: Computer Use tool type.
ComputerUseTool: A tool that allows the model to interact with the computer.
Condition: Filter condition applicable to a single key.
Content: Represents a message in a conversation (user or model turn).
ContentDeltaEvent: An event containing a delta update for content.
ContentEmbedding: An embedding for a piece of content.
ContentStartEvent: An event indicating that content output has started.
ContentStopEvent: An event indicating that content output has stopped.
ContextWindowCompressionConfig: Configuration for context window compression.
Corpus: A collection of Documents.
CountTokensRequest: Request to count tokens.
CountTokensResponse: Response from counting tokens.
CreateAgentInteractionParams: Parameters for creating an interaction with an agent.
CreateAuthTokenRequest: Request wrapper for creating an ephemeral authentication token.
CreateModelInteractionParams: Parameters for creating an interaction with a model.
Dataset: Reference to a training dataset.
DeepResearchAgentConfig: Configuration for the Deep Research agent.
Document: A collection of related content in a corpus.
DocumentContent: A document content block.
DocumentDelta: A document delta update.
DynamicAgentConfig: Configuration for dynamic agents.
DynamicRetrievalConfig: Describes the options to customize dynamic retrieval.
EmbedContentBatch: A resource representing a batch of EmbedContent requests.
EmbedContentBatchOutput: The output of an EmbedContent batch request.
EmbedContentBatchStats: Stats about an embed batch operation.
EmbedContentRequest: Request to embed content.
EmbedContentResponse: Response from embedding content.
EphemeralTokenCredentials: Ephemeral token authentication credentials.
EphemeralTokenProvider: Provider for ephemeral token authentication.
ErrorEvent: An event indicating that an error has occurred.
ExecutableCode: Code generated by the model that is meant to be executed.
ExecutableCodePart: Code for model to execute.
File: Represents an uploaded file.
FileData: Reference to an uploaded file (via /v1beta/files).
FileDataPart: Reference to uploaded file.
FileSearch: The FileSearch tool that retrieves knowledge from Semantic Retrieval corpora.
FileSearchCustomMetadata: User-provided custom metadata stored as key-value pairs for file search.
FileSearchResult: A File Search result item.
FileSearchResultContent: A File Search result content block.
FileSearchResultDelta: A File Search result delta update.
FileSearchStore: A FileSearchStore is a collection of Documents.
FileSearchTool: A tool that allows the model to search files.
FunctionCall: A predicted function call returned from the model.
FunctionCallContent: A function tool call content block.
FunctionCallDelta: A function call delta update.
FunctionCallingConfig: Configuration for specifying function calling behavior.
FunctionCallPart: Model's request to call a function.
FunctionDeclaration: Structured representation of a function declaration.
FunctionResponse: The result output from a function call.
FunctionResponseBlob: Raw media bytes for function response.
FunctionResponseInlinePart: A datatype containing media that is part of a FunctionResponse message.
FunctionResponsePart: Result from function execution.
FunctionResultContent: A function tool result content block.
FunctionResultDelta: A function result delta update.
FunctionTool: A function tool that can be called by the model.
GenerateAnswerRequest: Request to generate a grounded answer from the Model.
GenerateAnswerResponse: Response from the model for a grounded answer.
GenerateContentBatch: A resource representing a batch of GenerateContent requests.
GenerateContentBatchOutput: The output of a GenerateContent batch request.
GenerateContentRequest: Request to generate content.
GenerateContentResponse: Response from generating content.
GeneratedFile: A file generated on behalf of a user (e.g., video generated by Veo).
GenerationConfig: Configuration options for model generation.
GoAway: Go away message indicating session is ending.
GoogleAIClient: Main client for the GoogleAI API.
GoogleAIConfig: Configuration for the GoogleAI client.
GoogleMaps: The GoogleMaps Tool that provides geospatial context for the user's query.
GoogleSearch: Google Search tool configuration.
GoogleSearchCallContent: A Google Search call content block.
GoogleSearchCallDelta: A Google Search call delta update.
GoogleSearchResult: A Google Search result item.
GoogleSearchResultContent: A Google Search result content block.
GoogleSearchResultDelta: A Google Search result delta update.
GoogleSearchRetrieval: Tool to retrieve public web data for grounding, powered by Google.
GoogleSearchTool: A tool that allows the model to search Google.
GroundingAttribution: Attribution for a source that contributed to an answer.
GroundingChunk: Grounding chunk.
GroundingChunkCustomMetadata: Custom metadata associated with a grounding chunk.
GroundingChunkStringList: A list of strings associated with a grounding chunk.
GroundingMetadata: Metadata returned to client when grounding is enabled.
GroundingPassage: Passage included inline with a grounding configuration.
GroundingPassageId: Identifier for a passage within a GroundingPassage.
GroundingPassages: A repeated list of passages.
GroundingSupport: Grounding support.
Hyperparameters: Hyperparameters for tuning a model.
Image: An image result from a web search.
ImageConfig: Config for image generation features.
ImageContent: An image content block.
ImageDelta: An image delta update.
ImageSearch: Image search configuration.
ImportFileRequest: Request for ImportFile to import a File API file with a FileSearchStore.
ImportFileResponse: Response for ImportFile to import a File API file with a FileSearchStore.
InlineDataPart: Inline binary data (base64).
InlinedEmbedContentRequest: A single embed request in a batch with optional metadata.
InlinedEmbedContentRequests: A list of inlined embed requests for batch processing.
InlinedEmbedContentResponse: A single embed response in a batch with optional metadata and error.
InlinedEmbedContentResponses: A list of inlined embed responses from batch processing.
InlinedRequest: A single request in a batch with optional metadata.
InlinedRequests: A list of inlined requests for batch processing.
InlinedResponse: A single response in a batch with optional metadata and error.
InlinedResponses: A list of inlined responses from batch processing.
InputConfig: Configures the input to a batch request.
InputEmbedContentConfig: Configures the input to an embed batch request.
InputFeedback: Feedback related to the input data used to answer the question, as opposed to the model-generated response to the question.
Interaction: The Interaction resource.
InteractionCompleteEvent: An event indicating that an interaction has completed.
InteractionContent: The content of an interaction response.
InteractionDelta: A delta update for interaction content.
InteractionError: An error from an interaction.
InteractionEvent: An event from the Interactions API streaming response.
InteractionGenerationConfig: Configuration parameters for model interactions.
InteractionSpeechConfig: Configuration for speech interaction.
InteractionStartEvent: An event indicating that an interaction has started.
InteractionStatusUpdateEvent: An event indicating that an interaction's status has been updated.
InteractionTool: A tool that can be used by the model.
InteractionToolChoice: Configuration for tool choice.
InteractionUsage: Statistics on the interaction request's token usage.
Interval: Represents a time interval.
LatLng: An object that represents a latitude/longitude pair.
ListBatchesResponse: Response message for listing batch operations.
ListCachedContentsResponse: Response from listing cached contents.
ListCorporaResponse: Response from ListCorpora containing a paginated list of Corpora.
ListDocumentsResponse: Response from ListDocuments containing a paginated list of Documents.
ListFileSearchStoresResponse: Response from ListFileSearchStores containing a paginated list of FileSearchStores.
ListFilesResponse: Response from listing files.
ListGeneratedFilesResponse: Response for ListGeneratedFiles.
ListModelsResponse: Response from listing models.
ListOperationsResponse: Response from listing operations.
ListPermissionsResponse: Response message for listing permissions.
ListTunedModelsResponse: Response from listing tuned models.
LiveClient: Client for the Gemini Live API.
LiveConfig: Main configuration for Live API sessions.
LiveGenerationConfig: Generation configuration for Live API sessions.
LiveSession: A live streaming session with the Gemini Live API.
LiveVoices: Available prebuilt voices for Live API.
LogprobCandidate: A single candidate token with its log probability.
LogprobsResult: Token-level log probabilities.
LROPoller<T>: Poller for long-running operations (LRO).
Maps: A grounding chunk from Google Maps.
McpServer: A MCPServer is a server that can be called by the model to perform actions.
McpServerTool: A tool that allows the model to call an MCP (Model Context Protocol) server.
McpServerToolCallContent: An MCP Server tool call content block.
McpServerToolCallDelta: An MCP Server tool call delta update.
McpServerToolResultContent: An MCP Server tool result content block.
McpServerToolResultDelta: An MCP Server tool result delta update.
Media: A proto encapsulate various types of media.
MediaResolution: Media resolution for a media part.
MetadataFilter: User provided filter to limit retrieval based on Chunk or Document level metadata values.
ModalityTokenCount: Represents token counting info for a single modality.
ModalityTokens: Token count for a single response modality.
Model: Represents a model available in the GoogleAI API.
ModelStatus: Status of a model.
NoAuthCredentials: No authentication credentials.
NoAuthProvider: Provider that returns no authentication credentials.
Operation: Represents a long-running operation.
OperationError: Error information for a failed operation.
PaginatedResponse<T>: Generic paginated response.
Paginator<T>: Helper for paginated API responses.
Part: A single unit of content (text, media, function call, etc.).
PartMetadataPart: Custom metadata.
Permission: Permission resource grants user, group or everyone access to a resource.
PlaceAnswerSources: Collection of sources that provide answers about the features of a given place in Google Maps.
PrebuiltVoiceConfig: Configuration for prebuilt voices.
PredictLongRunningGeneratedVideoResponse: Veo response containing generated video samples.
PredictLongRunningMetadata: Metadata for PredictLongRunning long running operations.
PredictLongRunningOperation: Represents a long-running operation for video prediction.
PredictLongRunningRequest: Request message for PredictLongRunning operation.
PredictLongRunningResponse: Response message for PredictLongRunning operation.
PredictRequest: Request message for Predict operation.
PredictResponse: Response message for Predict operation.
ProactivityConfig: Configuration for proactive audio responses.
PromptFeedback: Feedback on the prompt in the request.
RealtimeInputConfig: Configuration for real-time input handling.
RegisterFilesRequest: Request to register files.
RegisterFilesResponse: Response from registering files.
RequestMetadata: Metadata about the HTTP request that caused an error.
ResponseMetadata: Metadata about the HTTP response that caused an error.
RetrievalConfig: Retrieval config.
RetrievalMetadata: Metadata related to retrieval in the grounding flow.
RetrievedContext: Chunk from context retrieved by the file search tool.
RetryPolicy: Retry policy configuration.
ReviewSnippet: Encapsulates a snippet of a user review that answers a question about the features of a specific place in Google Maps.
SafetyRating: Safety rating for a piece of content.
SafetySetting: Safety setting to block content at different thresholds.
Schema: Schema for structured data.
SearchEntryPoint: Google search entry point.
SearchTypes: Search types configuration.
Segment: Segment of the content.
SemanticRetrieverChunk: Identifier for a Chunk retrieved via Semantic Retriever specified in the GenerateAnswerRequest using SemanticRetrieverConfig.
SemanticRetrieverConfig: Configuration for retrieving grounding content from a Corpus or Document created using the Semantic Retriever API.
SessionResumptionConfig: Configuration for session resumption.
SessionResumptionUpdate: Session resumption update message.
SlidingWindow: Sliding window configuration for context compression.
SpeechConfig: Configuration for speech synthesis.
Status: Error status information for RPC APIs.
StreamableHttpTransport: A transport that can stream HTTP requests and responses.
TextContent: A text content block.
TextDelta: A text delta update.
TextPart: Text content.
ThinkingConfig: Configuration for thinking features.
ThoughtContent: A thought content block.
ThoughtPart: Reasoning step indicator.
ThoughtSignatureDelta: A thought signature delta update.
ThoughtSignaturePart: Cached thought key (base64).
ThoughtSummaryDelta: A thought summary delta update.
Tool: Tool that the model may use to generate a response.
ToolConfig: The Tool configuration containing parameters for specifying Tool use in the request.
TopCandidates: Top candidates for a token position.
Transcription: Transcription of audio content.
TransferOwnershipRequest: Request to transfer ownership of a resource.
TunedModel: A fine-tuned model.
TuningExample: A single training example.
TuningExamples: Training examples for tuning.
TuningTask: Record of a tuning task.
Turn: A conversation turn in an interaction.
UnknownClientMessage: Unknown client message type.
UnknownServerMessage: Unknown server message type.
UploadToFileSearchStoreRequest: Request for UploadToFileSearchStore.
UploadToFileSearchStoreResponse: Response from UploadToFileSearchStore.
UrlContextCallContent: A URL context call content block.
UrlContextCallDelta: A URL context call delta update.
UrlContextResult: A URL context result item.
UrlContextResultContent: A URL context result content block.
UrlContextResultDelta: A URL context result delta update.
UrlContextTool: A tool that allows the model to fetch URL context.
UsageMetadata: Token usage metadata for the request/response.
Video: Representation of a video.
VideoContent: A video content block.
VideoDelta: A video delta update.
VideoMetadata: Metadata for a video file.
VideoMetadataPart: Video timing/sampling metadata.
VoiceConfig: Configuration for voice selection.
Web: Chunk from the web.
WebSearch: Web search configuration.
WhiteSpaceConfig: Configuration for a white space chunking algorithm.

Enums

AbortionStage: Stage at which request was aborted.
ActivityHandling: How the model handles user activity during generation.
AnswerStyle: Style in which answers should be returned.
ApiMode: API mode determining which Google AI service to use.
ApiVersion: API version for the GoogleAI and Vertex AI APIs. https://ai.google.dev/gemini-api/docs/api-versions
AuthPlacement: Authentication placement strategy.
BatchState: State of a batch operation.
BlockReason: Specifies the reason why the input was blocked.
DocumentState: State of a Document.
EndSensitivity: Sensitivity for detecting the end of user speech.
EphemeralTokenPlacement: Ephemeral token placement strategy.
FileSource: Source of a file.
FileState: State of an uploaded file.
FinishReason: Why generation stopped.
FunctionCallingMode: Function calling mode.
FunctionResponseScheduling: Scheduling options for function responses.
GeneratedFileState: The state of a GeneratedFile.
GranteeType: The type of grantee for a permission.
HarmBlockThreshold: Threshold for blocking harmful content.
HarmCategory: Harm categories that can be blocked.
HarmProbability: The probability that content is harmful.
InteractionResponseModality: The modality of a response.
InteractionStatus: The status of an interaction.
InteractionThinkingLevel: Controls the level of thought tokens the model should generate.
InteractionThinkingSummaries: Controls whether to include thought summaries in the response.
InteractionToolChoiceType: The type of tool choice.
Language: Code execution language.
MediaResolutionLevel: Media resolution level.
ModelStage: Lifecycle stage of a model.
Outcome: Outcome of the code execution.
PermissionRole: The role granted by a permission.
SchemaType: Schema type.
StartSensitivity: Sensitivity for detecting the start of user speech.
TaskType: Type of task for which the embedding will be used.
ThinkingLevel: Controls the depth of reasoning before generating responses.
TunedModelState: State of a tuned model.
TurnCoverage: How much input is included in a turn.

Extensions

CandidateExtensions on Candidate: Convenience extensions for Candidate.
ContentExtensions on Content: Convenience extensions for Content.
GenerateContentResponseExtensions on GenerateContentResponse: Convenience extensions for GenerateContentResponse.
InteractionExtensions on Interaction: Convenience extensions for Interaction.

Functions

answerStyleFromString(String? value) → AnswerStyle: Converts string to AnswerStyle enum.
answerStyleToString(AnswerStyle style) → String: Converts AnswerStyle enum to string.
batchStateFromString(String? value) → BatchState: Converts a string to a BatchState.
batchStateToString(BatchState state) → String: Converts a BatchState to a string.
blockReasonFromString(String? value) → BlockReason: Converts string to BlockReason enum.
blockReasonToString(BlockReason reason) → String: Converts BlockReason enum to string.
fileSourceFromString(String? value) → FileSource: Converts a string to a FileSource.
fileSourceToString(FileSource source) → String: Converts a FileSource to a string.
fileStateFromString(String? value) → FileState: Converts a string to a FileState.
fileStateToString(FileState state) → String: Converts a FileState to a string.
finishReasonFromString(String? value) → FinishReason: Converts string to FinishReason enum.
finishReasonToString(FinishReason reason) → String: Converts FinishReason enum to string.
functionCallingModeFromString(String? value) → FunctionCallingMode: Converts a string to a FunctionCallingMode.
functionCallingModeToString(FunctionCallingMode mode) → String: Converts a FunctionCallingMode to a string.
generatedFileStateFromString(String? value) → GeneratedFileState: Converts a string to a GeneratedFileState.
generatedFileStateToString(GeneratedFileState? state) → String: Converts a GeneratedFileState to a string.
granteeTypeFromString(String? value) → GranteeType: Converts a string to GranteeType.
granteeTypeToString(GranteeType type) → String: Converts GranteeType to a string.
harmBlockThresholdFromString(String? value) → HarmBlockThreshold: Converts string to HarmBlockThreshold enum.
harmBlockThresholdToString(HarmBlockThreshold threshold) → String: Converts HarmBlockThreshold enum to string.
harmCategoryFromString(String? value) → HarmCategory: Converts string to HarmCategory enum.
harmCategoryToString(HarmCategory category) → String: Converts HarmCategory enum to string.
harmProbabilityFromString(String? value) → HarmProbability: Converts string to HarmProbability enum.
harmProbabilityToString(HarmProbability probability) → String: Converts HarmProbability enum to string.
interactionResponseModalityFromString(String? value) → InteractionResponseModality: Converts a string to InteractionResponseModality.
interactionResponseModalityToString(InteractionResponseModality modality) → String: Converts InteractionResponseModality to a string.
interactionStatusFromString(String? value) → InteractionStatus: Converts a string to InteractionStatus.
interactionStatusToString(InteractionStatus status) → String: Converts InteractionStatus to a string.
interactionThinkingLevelFromString(String? value) → InteractionThinkingLevel: Converts a string to InteractionThinkingLevel.
interactionThinkingLevelToString(InteractionThinkingLevel level) → String: Converts InteractionThinkingLevel to a string.
interactionThinkingSummariesFromString(String? value) → InteractionThinkingSummaries: Converts a string to InteractionThinkingSummaries.
interactionThinkingSummariesToString(InteractionThinkingSummaries summaries) → String: Converts InteractionThinkingSummaries to a string.
interactionToolChoiceTypeFromString(String? value) → InteractionToolChoiceType: Converts a string to InteractionToolChoiceType.
interactionToolChoiceTypeToString(InteractionToolChoiceType type) → String: Converts InteractionToolChoiceType to a string.
modelStageFromString(String? value) → ModelStage: Parses a ModelStage from its string representation.
modelStageToString(ModelStage value) → String: Converts a ModelStage to its string representation.
permissionRoleFromString(String? value) → PermissionRole: Converts a string to PermissionRole.
permissionRoleToString(PermissionRole role) → String: Converts PermissionRole to a string.
schemaTypeFromString(String? value) → SchemaType: Converts string to SchemaType enum.
schemaTypeToString(SchemaType type) → String: Converts SchemaType enum to string.
taskTypeFromString(String? value) → TaskType: Converts string to TaskType enum.
taskTypeToString(TaskType type) → String: Converts TaskType enum to string.
thinkingLevelFromString(String? value) → ThinkingLevel: Converts string to ThinkingLevel enum.
thinkingLevelToString(ThinkingLevel level) → String: Converts ThinkingLevel enum to string.
tunedModelStateFromString(String? value) → TunedModelState: Converts a string to a TunedModelState.
tunedModelStateToString(TunedModelState state) → String: Converts a TunedModelState to a string.

Exceptions / Errors

AbortedException: Exception thrown when a request is aborted by user.
ApiException: Exception for HTTP/API errors.
AuthenticationException: Exception for authentication errors (HTTP 401).
GoogleAIException: Base sealed class for all GoogleAI exceptions.
LiveConnectionException: Exception thrown when WebSocket connection fails.
LiveSessionClosedException: Exception thrown when the Live session is closed.
LiveSessionException: Exception thrown for general Live session errors.
LiveSessionResumptionException: Exception thrown when session resumption fails.
LiveSessionSetupException: Exception thrown when session setup fails.
RateLimitException: Exception for rate limit (429) errors.
StreamException: Exception for errors detected in streaming data.
TimeoutException: Exception for request timeouts.
ValidationException: Exception for client-side validation errors.

googleai_dart library

Classes

Enums

Extensions

Functions

Exceptions / Errors

googleai_dart package

googleai_dart library