models/tool_models library

Classes

AnyToolChoice: Model can use any tool, but it must use at least one. This is useful when you want to force the model to use tools.
AutoToolChoice: Model can use any tool, and may elect to use none. This is the default behavior and gives the model flexibility.
FunctionObject: Represents a function object for assistants (similar to FunctionTool but with optional parameters)
FunctionTool: Represents a function definition for a tool
NoneToolChoice: Explicitly disables the use of tools. The model will not use any tools even if they are provided.
ParallelToolConfig: Parallel tool execution configuration
ParameterProperty: Represents a parameter in a function tool
ParametersSchema: Represents the parameters schema for a function tool
SpecificToolChoice: Model must use the specified tool and only the specified tool. The string parameter is the name of the required tool. This is useful when you want the model to call a specific function.
StructuredOutputFormat: Defines rules for structured output responses based on OpenAI's structured output requirements.
Tool: Represents a tool that can be used in chat
ToolChoice: Tool choice determines how the LLM uses available tools. The behavior is standardized across different LLM providers.
ToolResult: Tool execution result that can be returned to the model