Struct LLMBuilder

pub struct LLMBuilder<L: LLMProvider> { /* private fields */ }

Builder for configuring and instantiating LLM providers.

Provides a fluent interface for setting various configuration options like model selection, API keys, generation parameters, etc.

Implementations

impl<L: LLMProvider> LLMBuilder<L>

pub fn new() -> Self

Creates a new empty builder instance with default values.

pub fn api_key(self, key: impl Into<String>) -> Self

Sets the API key for authentication.

pub fn base_url(self, url: impl Into<String>) -> Self

Sets the base URL for API requests.

pub fn model(self, model: impl Into<String>) -> Self

Sets the model identifier to use.

pub fn max_tokens(self, max_tokens: u32) -> Self

Sets the maximum number of tokens to generate.

pub fn temperature(self, temperature: f32) -> Self

Sets the temperature for controlling response randomness (0.0-1.0).

pub fn tools(self, tools: Vec<Tool>) -> Self

pub fn system(self, system: impl Into<String>) -> Self

Sets the system prompt/context.

pub fn reasoning_effort(self, reasoning_effort: ReasoningEffort) -> Self

Sets the reasoning flag.

pub fn reasoning(self, reasoning: bool) -> Self

Sets the reasoning flag.

pub fn reasoning_budget_tokens(self, reasoning_budget_tokens: u32) -> Self

Sets the reasoning budget tokens.

pub fn timeout_seconds(self, timeout_seconds: u64) -> Self

Sets the request timeout in seconds.

pub fn stream(self, stream: bool) -> Self

Enables or disables streaming responses.

pub fn top_p(self, top_p: f32) -> Self

Sets the top-p (nucleus) sampling parameter.

pub fn top_k(self, top_k: u32) -> Self

Sets the top-k sampling parameter.

pub fn embedding_encoding_format( self, embedding_encoding_format: impl Into<String>, ) -> Self

Sets the encoding format for embeddings.

pub fn embedding_dimensions(self, embedding_dimensions: u32) -> Self

Sets the dimensions for embeddings.

pub fn schema(self, schema: impl Into<StructuredOutputFormat>) -> Self

Sets the JSON schema for structured output.

pub fn validator<F>(self, f: F) -> Self

where F: Fn(&str) -> Result<(), String> + Send + Sync + 'static,

Sets a validation function to verify LLM responses.

Arguments

• f - Function that takes a response string and returns Ok(()) if valid, or Err with error message if invalid

pub fn validator_attempts(self, attempts: usize) -> Self

Sets the number of retry attempts for validation failures.

Arguments

• attempts - Maximum number of times to retry generating a valid response

pub fn function(self, function_builder: FunctionBuilder) -> Self

Adds a function tool to the builder

pub fn enable_parallel_tool_use(self, enable: bool) -> Self

Enable parallel tool use

pub fn tool_choice(self, choice: ToolChoice) -> Self

Set tool choice. Note that if the choice is given as Tool(name), and that tool isn’t available, the builder will fail.

pub fn disable_tools(self) -> Self

Explicitly disable the use of tools, even if they are provided.

pub fn api_version(self, api_version: impl Into<String>) -> Self

Set the API version.

pub fn deployment_id(self, deployment_id: impl Into<String>) -> Self

Set the deployment id. Used in Azure OpenAI.

Trait Implementations

impl<L: LLMProvider> Default for LLMBuilder<L>

fn default() -> Self

Returns the “default value” for a type.