Struct OpenAI 

pub struct OpenAI { /* private fields */ }

An OpenAI-shaped provider: its OpenAIConfig on a transport. Every model it builds sends through that transport.

use rig_core::completion::CompletionRequest;
use rig_core::providers::openai::{self, OpenAIConfig};

let openai = OpenAIConfig::from_env()?.connect(http);
let model = openai.completion(openai::GPT_5_2);
let response = model.call(CompletionRequest::new("Capital of France?")).await?;

Implementations

impl OpenAI

pub fn with_http(self, http: impl HttpClientExt + 'static) -> Self

The same configuration, sending through http.

pub fn config(&self) -> &OpenAIConfig

The configuration this client sends with.

impl OpenAI

pub fn new(api_key: impl Into<Secret>) -> Self

Available on crate feature reqwest only.

Official OpenAI with api_key, on the shared reqwest client.

pub fn from_env() -> Result<Self, EnvError>

Available on crate feature reqwest only.

Official OpenAI from OPENAI_API_KEY and the optional OPENAI_BASE_URL, on the shared reqwest client.

pub fn completion(&self, model: impl Into<String>) -> Model<OpenAiWire>

The completion model for model on the configuration’s completion_route.

pub fn responses(&self, model: impl Into<String>) -> Model<Responses>

The Responses model for model: POST /responses.

pub fn chat(&self, model: impl Into<String>) -> Model<Chat>

The Chat Completions model for model.

pub fn embedding( &self, model: impl Into<String>, ndims: Option<usize>, ) -> Model<Embeddings>

The embedding model for model, ndims wide when set.

pub fn rerank(&self, model: impl Into<String>) -> Model<Rerank>

The rerank model for model.

pub fn transcription(&self, model: impl Into<String>) -> Model<Transcriptions>

The transcription model for model.

pub fn image_generation(&self, model: impl Into<String>) -> Model<Images>

Available on crate feature image only.

The image-generation model for model.

pub fn audio_generation(&self, model: impl Into<String>) -> Model<Speech>

Available on crate feature audio only.

The speech model for model.

pub async fn list_models(&self) -> Result<ModelList, ProviderError>

The models this provider serves, every page followed.

pub async fn verify(&self) -> Result<(), ProviderError>

Check that the provider accepts the configured credential. A 401 or 403 reply is ProviderError::InvalidAuthentication.

pub async fn authenticate( self, authenticator: &Authenticator, ) -> Result<Self, AuthError>

This client with the ChatGPT subscription credential authenticator resolves, reading or refreshing it through this client’s transport: the access token, and the account it belongs to when one is named. Every other setting is kept.

use rig_core::providers::chatgpt::{self, auth::{AuthSource, Authenticator, DeviceCodeHandler}};
use rig_core::providers::openai::OpenAIConfig;

let authenticator = Authenticator::new(AuthSource::OAuth, None, DeviceCodeHandler::default(), true);
let chatgpt = OpenAIConfig::with_key(&chatgpt::DIALECT, "")
    .connect(http)
    .authenticate(&authenticator)
    .await?;

Trait Implementations

impl Clone for OpenAI

fn clone(&self) -> Self

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for OpenAI

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

Formats the value using the given formatter.