Struct OpenAIClient 

pub struct OpenAIClient { /* private fields */ }

Async LLM client for the OpenAI chat completions API.

Implementations

impl OpenAIClient

pub fn new(config: &ModelConfig) -> Result<Self>

Create a new client using credentials from the environment variable in config.

pub fn from_key( model: impl Into<String>, api_key: impl Into<String>, ) -> Result<Self>

Create a new client with an explicit API key, using the default OpenAI base URL.

Returns a KernelError::Config if the HTTP client (with its connect / total timeouts) cannot be built, rather than silently falling back to a timeout-less reqwest::Client::default().

Example

use llm_kernel::llm::OpenAIClient;
let client = OpenAIClient::from_key("gpt-4o-mini", "sk-...")?;

pub fn from_key_with_client( model: impl Into<String>, api_key: impl Into<String>, client: Client, ) -> Self

Create from an explicit key and a shared reqwest::Client.

Prefer this over from_key when constructing multiple clients in a hot path — the shared client reuses the underlying TCP connection pool.

pub fn from_key_with_base_url( model: impl Into<String>, api_key: impl Into<String>, base_url: impl Into<String>, client: Client, ) -> Self

Create from an explicit key, a custom base URL, and a shared reqwest::Client.

Use this for OpenAI-compatible providers that are not the default OpenAI endpoint (DeepSeek, Groq, Ollama, LM Studio, custom gateways, …) when you already hold the key in memory and want to reuse a shared connection pool.

Trait Implementations

impl LLMClient for OpenAIClient

fn complete<'life0, 'async_trait>( &'life0 self, request: LLMRequest, ) -> Pin<Box<dyn Future<Output = Result<LLMResponse>> + Send + 'async_trait>>

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

Send a chat completion request and return the full response.

fn model_name(&self) -> &str

Return the model name this client is configured to use.

fn stream_complete<'life0, 'async_trait>( &'life0 self, request: LLMRequest, ) -> Pin<Box<dyn Future<Output = Result<LLMStream>> + Send + 'async_trait>>

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

Stream a chat completion, yielding events as they arrive.