Struct HttpOllamaClient 

pub struct HttpOllamaClient { /* private fields */ }

Production client that speaks to a real Ollama server over HTTP.

Implementations

impl HttpOllamaClient

pub fn new(base_url: impl Into<String>) -> Self

base_url is typically "http://localhost:11434".

Trait Implementations

impl OllamaClient for HttpOllamaClient

fn generate<'life0, 'async_trait>( &'life0 self, req: GenerateRequest, ) -> Pin<Box<dyn Future<Output = Result<GenerateResponse, OxideError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Single-turn text completion (/api/generate).

fn chat<'life0, 'async_trait>( &'life0 self, req: ChatRequest, ) -> Pin<Box<dyn Future<Output = Result<ChatResponse, OxideError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Multi-turn chat completion (/api/chat), including tool/function calling.

fn embed<'life0, 'async_trait>( &'life0 self, req: EmbedRequest, ) -> Pin<Box<dyn Future<Output = Result<EmbedResponse, OxideError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Produce dense vector embeddings for one or more texts (/api/embed).

fn list_models<'life0, 'async_trait>( &'life0 self, ) -> Pin<Box<dyn Future<Output = Result<ListModelsResponse, OxideError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

List models available on the Ollama server (/api/tags).

fn stream_generate(&self, req: GenerateRequest) -> BoxStream<GenerateResponse>

Stream a generate response token-by-token.

fn stream_chat(&self, req: ChatRequest) -> BoxStream<ChatResponse>

Stream a chat response token-by-token.