google_generative_ai library
Google generative AI SDK API bindings for Dart.
The Google Generative AI SDK for Dart allows developers to use state-of-the-art Large Language Models (LLMs) to build applications.
Most uses of this library will be through a GenerativeModel. Here's a simple example of using this API:
const apiKey = ...;
void main() async {
final model = GenerativeModel(
model: 'gemini-1.5-flash-latest',
apiKey: apiKey,
);
final prompt = 'Write a story about a magic backpack.';
final content = [Content.text(prompt)];
final response = await model.generateContent(content);
print(response.text);
};
Classes
- BatchEmbedContentsResponse
- Candidate
- Response candidate generated from a GenerativeModel.
- ChatSession
- A back-and-forth chat with a generative model.
- CitationMetadata
- Source attributions for a piece of content.
- CitationSource
- Citation to a source for a portion of a specific response.
- Content
- The base structured datatype containing multi-part content of a message.
- ContentEmbedding
- An embedding, as defined by a list of values.
- CountTokensResponse
- DataPart
- A Part with the byte content of a file.
- EmbedContentRequest
- EmbedContentResponse
- FilePart
- A Part referring to an uploaded file.
- FunctionCall
-
A predicted
FunctionCall
returned from the model that contains a string representing theFunctionDeclaration.name
with the arguments and their values. - FunctionCallingConfig
- Configuration specifying how the model should use the functions provided as tools.
- FunctionDeclaration
- Structured representation of a function declaration as defined by the OpenAPI 3.03 specification.
- FunctionResponse
- GenerateContentResponse
- Response from the model; supports multiple candidates.
- GenerationConfig
- Configuration options for model generation and outputs.
- GenerativeModel
- A multimodel generative model (like Gemini).
- Part
- A datatype containing media that is part of a multi-part Content message.
- PromptFeedback
- Feedback metadata of a prompt specified in a GenerativeModel request.
- RequestOptions
- Configuration for how a GenerativeModel makes requests.
- SafetyRating
- Safety rating for a piece of content.
- SafetySetting
- Safety setting, affecting the safety-blocking behavior.
- Schema
- The definition of an input or output data types.
- TextPart
- Tool
- Tool details that the model may use to generate a response.
- ToolConfig
- UsageMetadata
- Metadata on the generation request's token usage.
Enums
- BlockReason
- The reason why a prompt was blocked.
- FinishReason
- Reason why a model stopped generating tokens.
- FunctionCallingMode
- HarmBlockThreshold
- Probability of harm which causes content to be blocked.
- HarmCategory
- The category of a rating.
- HarmProbability
- The probability that a piece of content is harmful.
- SchemaType
- The value type of a Schema.
- TaskType
- Type of task for which the embedding will be used.
Extensions
Exceptions / Errors
- GenerativeAIException
- Exception thrown when generating content fails.
- InvalidApiKey
- Exception thrown when the server rejects the API key.
- ServerException
- Exception thrown when the server failed to generate content.
- UnsupportedUserLocation
- Exception thrown when the user location is unsupported.