LlmContext class

Manages the execution context for LLM requests

Constructors

LlmContext({Map<String, dynamic> contextInfo = const {}, String? systemInstructions, int? maxInputTokens, int? maxOutputTokens, Duration? timeout, bool allowToolUse = true, List<String>? allowedTools, double temperature = 0.7, double? topP, int? topK})

const

Properties

allowedTools → List<String>?

final

allowToolUse → bool

final

contextInfo → Map<String, dynamic>

Additional context information to guide LLM responses

final

hashCode → int

The hash code for this object.

no setter, inherited

maxInputTokens → int?

Context window sizing parameters

final

maxOutputTokens → int?

final

runtimeType → Type

A representation of the runtime type of the object.

no setter, inherited

systemInstructions → String?

System instructions to shape LLM behavior

final

temperature → double

Execution preferences

final

timeout → Duration?

Execution constraints

final

topK → int?

final

topP → double?

final

Methods

copyWith({Map<String, dynamic>? contextInfo, String? systemInstructions, int? maxInputTokens, int? maxOutputTokens, Duration? timeout, bool? allowToolUse, List<String>? allowedTools, double? temperature, double? topP, int? topK}) → LlmContext

Create a new context with updated parameters

noSuchMethod(Invocation invocation) → dynamic

Invoked when a nonexistent method or property is accessed.

inherited

toJson() → Map<String, dynamic>

Convert the LLM context to a JSON representation

toProviderParams(String provider) → Map<String, dynamic>

Convert to parameter map for LLM provider

toString() → String

A string representation of this object.

inherited

Operators

operator ==(Object other) → bool

The equality operator.

inherited

Static Methods

code() → LlmContext

Create a context optimized for code generation

concise() → LlmContext

Create a context optimized for concise responses

creative() → LlmContext

Create a context optimized for creative generation

factual() → LlmContext

Create a context optimized for factual responses

merge(LlmContext base, LlmContext overlay) → LlmContext

Merge two contexts, with the second context taking precedence