Struct OpenAiProvider 

pub struct OpenAiProvider { /* private fields */ }

OpenAI provider implementing Provider.

Supports the OpenAI Chat Completions API with tool calling, structured output, and streaming.

Example

use llm_stack_openai::{OpenAiConfig, OpenAiProvider};
use llm_stack::{ChatParams, ChatMessage, Provider};

let provider = OpenAiProvider::new(OpenAiConfig {
    api_key: std::env::var("OPENAI_API_KEY").unwrap(),
    ..Default::default()
});

let response = provider.generate(&ChatParams {
    messages: vec![ChatMessage::user("Hello!")],
    ..Default::default()
}).await?;

Implementations

impl OpenAiProvider

pub fn new(config: OpenAiConfig) -> Self

Create a new OpenAI provider from configuration.

If config.client is Some, that client is reused for connection pooling. Otherwise a new client is built with the configured timeout.

Trait Implementations

impl Debug for OpenAiProvider

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Provider for OpenAiProvider

async fn generate(&self, params: &ChatParams) -> Result<ChatResponse, LlmError>

Sends a chat completion request and returns the full response.

async fn stream(&self, params: &ChatParams) -> Result<ChatStream, LlmError>

Sends a chat completion request and returns a stream of events.

fn metadata(&self) -> ProviderMetadata

Returns static metadata describing this provider instance.