Struct RetryClient 

pub struct RetryClient<C> { /* private fields */ }

An LLMClient wrapper that retries on transient errors.

Retries are attempted for:

• 429 Rate Limited — uses the retry-after header value as the initial delay.
• 5xx Server Errors — exponential backoff starting from base_delay.

All other errors (4xx, network timeouts, deserialization failures) are returned immediately without retry.

Implementations

impl<C> RetryClient<C>

pub fn new(inner: C, config: RetryConfig) -> Self

Wrap an LLMClient with retry behavior.

pub fn inner(&self) -> &C

Access the underlying client.

Trait Implementations

impl<C: LLMClient> LLMClient for RetryClient<C>

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,

Streaming is not retried. Delegates directly to the inner client.

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.