Struct OpenAI

pub struct OpenAI {

    pub api_key: String,
    pub model: String,
    pub max_tokens: Option<u32>,
    pub temperature: Option<f32>,
    pub system: Option<String>,
    pub timeout_seconds: Option<u64>,
    pub stream: Option<bool>,
    pub top_p: Option<f32>,
    pub top_k: Option<u32>,
    pub tools: Option<Vec<Tool>>,
    pub embedding_encoding_format: Option<String>,
    pub embedding_dimensions: Option<u32>,
    pub reasoning_effort: Option<String>,
    /* private fields */
}

Client for interacting with OpenAI’s API.

Provides methods for chat and completion requests using OpenAI’s models.

Fields

api_key: String

model: String

max_tokens: Option<u32>

temperature: Option<f32>

system: Option<String>

timeout_seconds: Option<u64>

stream: Option<bool>

top_p: Option<f32>

top_k: Option<u32>

tools: Option<Vec<Tool>>

embedding_encoding_format: Option<String>

Embedding parameters

embedding_dimensions: Option<u32>

reasoning_effort: Option<String>

Implementations

impl OpenAI

pub fn new( api_key: impl Into<String>, model: Option<String>, max_tokens: Option<u32>, temperature: Option<f32>, timeout_seconds: Option<u64>, system: Option<String>, stream: Option<bool>, top_p: Option<f32>, top_k: Option<u32>, embedding_encoding_format: Option<String>, embedding_dimensions: Option<u32>, tools: Option<Vec<Tool>>, reasoning_effort: Option<String>, ) -> Self

Creates a new OpenAI client with the specified configuration.

Arguments

• api_key - OpenAI API key
• model - Model to use (defaults to “gpt-3.5-turbo”)
• max_tokens - Maximum tokens to generate
• temperature - Sampling temperature
• timeout_seconds - Request timeout in seconds
• system - System prompt
• stream - Whether to stream responses

Trait Implementations

impl ChatProvider for OpenAI

fn chat_with_tools<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, messages: &'life1 [ChatMessage], tools: Option<&'life2 [Tool]>, ) -> Pin<Box<dyn Future<Output = Result<Box<dyn ChatResponse>, LLMError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Sends a chat request to OpenAI’s API.

Arguments

• messages - Slice of chat messages representing the conversation
• tools - Optional slice of tools to use in the chat

Returns

The model’s response text or an error

fn chat<'life0, 'life1, 'async_trait>( &'life0 self, messages: &'life1 [ChatMessage], ) -> Pin<Box<dyn Future<Output = Result<Box<dyn ChatResponse>, LLMError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Sends a chat request to the provider with a sequence of messages.

impl CompletionProvider for OpenAI

fn complete<'life0, 'life1, 'async_trait>( &'life0 self, _req: &'life1 CompletionRequest, ) -> Pin<Box<dyn Future<Output = Result<CompletionResponse, LLMError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Sends a completion request to OpenAI’s API.

Currently not implemented.

impl EmbeddingProvider for OpenAI

fn embed<'life0, 'async_trait>( &'life0 self, input: Vec<String>, ) -> Pin<Box<dyn Future<Output = Result<Vec<Vec<f32>>, LLMError>> + Send + 'async_trait>>

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

impl LLMProvider for OpenAI

fn tools(&self) -> Option<&[Tool]>