Struct OpenAiResponsesProvider 

pub struct OpenAiResponsesProvider { /* private fields */ }

The live OpenAI Responses Provider (non-streaming, stateless).

Implementations

impl OpenAiResponsesProvider

pub fn new( config: OpenAiModelConfig, ) -> Result<OpenAiResponsesProvider, ProviderError>

Build a provider from a resolved OpenAiModelConfig.

Errors

ProviderError::Transport when the HTTP client cannot be constructed.

pub fn from_env() -> Result<OpenAiResponsesProvider, ProviderError>

Build from the environment (LOCODE_API_KEY / LOCODE_BASE_URL / LOCODE_MODEL).

Errors

ProviderError::Auth when the key is missing; ProviderError::Transport when the client cannot be constructed.

pub fn with_retry_policy(self, retry: RetryPolicy) -> OpenAiResponsesProvider

Override the transport retry policy.

pub fn config(&self) -> &OpenAiModelConfig

The active config (read-only view).

pub fn config_mut(&mut self) -> &mut OpenAiModelConfig

Mutable config access (the facade sets prompt_cache_key to the session id, plan §A.5 Q4).

Trait Implementations

impl Provider for OpenAiResponsesProvider

fn api_schema(&self) -> &str

The wire-schema id this provider speaks — e.g. "anthropic", "mock".

fn complete<'life0, 'life1, 'async_trait>( &'life0 self, request: &'life1 ConversationRequest, ) -> Pin<Box<dyn Future<Output = Result<Completion, ProviderError>> + Send + 'async_trait>>

where 'life0: 'async_trait, 'life1: 'async_trait, OpenAiResponsesProvider: 'async_trait,

Sample one completion for request.