Struct LlmConfig 

pub struct LlmConfig {
    pub api_key: String,
    pub base_url: String,
    pub model: String,
    pub infer_retries: usize,
    pub temperature: Option<f64>,
    pub max_tokens: Option<i64>,
}

Configuration for the LLM client.

Fields

api_key: String

API key for authentication.

base_url: String

Base URL for the API.

model: String

Model to use.

infer_retries: usize

Max retries for structured inference.

temperature: Option<f64>

Temperature for sampling (0.0 - 2.0). None uses API default.

max_tokens: Option<i64>

Maximum tokens to generate. None uses API default.

Implementations

impl LlmConfig

pub fn from_env() -> Self

Create a config from environment variables (native) or WASM config injection (browser).

pub fn mock() -> Self

Create a mock config for testing.

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

Create a config with specific model and defaults for other settings.

This is useful when you want to override only specific fields like model from an effect handler, while keeping API key and base URL from environment.

pub fn with_temperature(self, temp: f64) -> Self

Set the temperature for this config.

pub fn with_max_tokens(self, tokens: i64) -> Self

Set the max tokens for this config.

pub fn is_mock(&self) -> bool

Check if this is a mock configuration.

pub fn is_ollama(&self) -> bool

Check if the base URL points to a local Ollama instance.

Trait Implementations

impl Clone for LlmConfig

fn clone(&self) -> LlmConfig

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.