LlmProvider

gestura_core_llm

Trait LlmProvider 

Source 
pub trait LlmProvider: Send + Sync {
    // Required method
    fn call<'life0, 'life1, 'async_trait>(
        &'life0 self,
        prompt: &'life1 str,
    ) -> Pin<Box<dyn Future<Output = Result<String, AppError>> + Send + 'async_trait>>
       where Self: 'async_trait,
             'life0: 'async_trait,
             'life1: 'async_trait;

    // Provided methods
    fn call_with_usage<'life0, 'life1, 'async_trait>(
        &'life0 self,
        prompt: &'life1 str,
    ) -> Pin<Box<dyn Future<Output = Result<LlmCallResponse, AppError>> + Send + 'async_trait>>
       where Self: 'async_trait,
             'life0: 'async_trait,
             'life1: 'async_trait { ... }
    fn call_with_tools<'life0, 'life1, 'life2, 'async_trait>(
        &'life0 self,
        prompt: &'life1 str,
        _tools: Option<&'life2 [Value]>,
    ) -> Pin<Box<dyn Future<Output = Result<LlmCallResponse, AppError>> + Send + 'async_trait>>
       where Self: 'async_trait,
             'life0: 'async_trait,
             'life1: 'async_trait,
             'life2: 'async_trait { ... }
}
Expand description

Unified LLM interface (async)

Required Methods§

Source

fn call<'life0, 'life1, 'async_trait>( &'life0 self, prompt: &'life1 str, ) -> Pin<Box<dyn Future<Output = Result<String, AppError>> + Send + 'async_trait>>
where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Call the LLM with a prompt and return the generated text. For backward compatibility, this returns just the text.

Provided Methods§

Source

fn call_with_usage<'life0, 'life1, 'async_trait>( &'life0 self, prompt: &'life1 str, ) -> Pin<Box<dyn Future<Output = Result<LlmCallResponse, AppError>> + Send + 'async_trait>>
where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Call the LLM with a prompt and return the response with token usage. Default implementation calls call and returns unknown usage.

Source

fn call_with_tools<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, prompt: &'life1 str, _tools: Option<&'life2 [Value]>, ) -> Pin<Box<dyn Future<Output = Result<LlmCallResponse, AppError>> + Send + 'async_trait>>
where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Call the LLM with a prompt and optional tool schemas.

When tools is Some, providers that support native tool/function calling will include the schemas in the API request body, enabling the model to return structured tool call responses.

The default implementation ignores the tools parameter and delegates to Self::call_with_usage. Providers should override this to pass tools natively.

Implementors§