LLMProvider class abstract

Abstract interface for LLM (Large Language Model) providers.

This interface defines the contract that all LLM providers must implement to be used with Spectra. Providers handle the actual communication with AI services (OpenAI, Anthropic, Google, etc.).

Implementing a Provider

To create a new provider, extend this class and implement:

• generateResponse - The main method for generating AI responses
• name - Human-readable name of the provider
• availableModels - List of supported model names
• defaultModel - The recommended default model

Example

class MyProvider implements LLMProvider {
  @override
  String get name => 'MyProvider';

  @override
  List<String> get availableModels => ['model-1', 'model-2'];

  @override
  String get defaultModel => 'model-1';

  @override
  Future<String> generateResponse(String prompt, {List<String>? context}) async {
    // Implementation here
  }
}

Implementers

• CachedLLMProvider
• ClaudeProvider
• DeepSeekProvider
• GeminiProvider
• GrokProvider
• OpenAIProvider

Constructors

LLMProvider()

Properties

availableModels → List<String>

List of model names available from this provider.

no setter

defaultModel → String

The default model to use if none is specified.

no setter

hashCode → int

The hash code for this object.

no setter, inherited

name → String

Human-readable name of this provider (e.g., "OpenAI", "Claude").

no setter

runtimeType → Type

A representation of the runtime type of the object.

no setter, inherited

Methods

generateResponse(String prompt, {List<String>? context}) → Future<String>

Generates a response from the LLM based on the given prompt.

noSuchMethod(Invocation invocation) → dynamic

Invoked when a nonexistent method or property is accessed.

inherited

toString() → String

A string representation of this object.

inherited

Operators

operator ==(Object other) → bool

The equality operator.

inherited