Enum LlmClientError 

#[non_exhaustive]
pub enum LlmClientError {
    InvalidRequest {
        message: String,
    },
    RequestTranslation(String),
    RequestEncoding(String),
    ResponseTranslation(String),
    Configuration {
        message: String,
    },
    ResponseStateLimitExceeded {
        limit: usize,
    },
    ResponseStateConflict,
    TemporarilyUnavailable,
    Transport {
        source: BoxError,
    },
    Timeout {
        source: BoxError,
    },
    ContextWindowExceeded {
        model: ModelId,
        message: String,
    },
    UpstreamHttp {
        status: StatusCode,
        body: String,
    },
    InvalidResponse {
        source: BoxError,
    },
    Host {
        source: BoxError,
    },
    Ffi {
        source: BoxError,
    },
    General(String),
}

Failures a routed LLM client can surface to its caller.

Boxed sources preserve client-specific errors; General keeps only a message.

Variants (Non-exhaustive)

Non-exhaustive enums could have additional variants added in future. Therefore, when matching against variants of non-exhaustive enums, an extra wildcard arm must be added to account for any future variants.

InvalidRequest

The request cannot be served as supplied.

Fields

message: String

Human-readable request validation failure.

RequestTranslation(String)

Decoding the inbound request failed in the translation engine.

RequestEncoding(String)

Encoding the request for the upstream failed in the translation engine.

ResponseTranslation(String)

Decoding or encoding the response failed in the translation engine.

Configuration

The client is not configured to serve the selected target.

Fields

message: String

Human-readable configuration failure.

ResponseStateLimitExceeded

The route cannot record another provider-owned response or conversation ID.

Fields

limit: usize

Maximum number of IDs retained by this route.

ResponseStateConflict

Two configured models reported the same saved response or conversation ID.

TemporarilyUnavailable

The selected backend is temporarily unavailable. Another candidate may serve the request.

Transport

The upstream could not be reached or the request could not be sent.

Fields

source: BoxError

Client-specific transport failure.

Timeout

The upstream request exceeded its timeout.

Fields

source: BoxError

Client-specific timeout failure.

ContextWindowExceeded

The upstream rejected the request because it exceeds the model’s context window.

Fields

model: ModelId

Model whose context window was exceeded.

message: String

Upstream error message.

UpstreamHttp

The upstream returned a non-success HTTP response.

Fields

status: StatusCode

Upstream HTTP status code.

body: String

Raw upstream error body.

InvalidResponse

The upstream returned a response the client could not decode.

Fields

source: BoxError

Client-specific decoding or validation failure.

Host

A native host failure that does not fit another category. The default completion runner treats it as terminal.

Fields

source: BoxError

Original error, retained for the host.

Ffi

A call across a foreign-function boundary (e.g. a Python-implemented client) failed. The boxed source is the foreign error itself.

Fields

source: BoxError

Foreign-language failure, preserved verbatim.

General(String)

A string-only failure. Useful in testing; prefer a typed variant when possible.

Trait Implementations

impl Debug for LlmClientError

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

Formats the value using the given formatter.

impl Display for LlmClientError

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

Formats the value using the given formatter.

impl Error for LlmClientError

fn source(&self) -> Option<&(dyn Error + 'static)>

Returns the lower-level source of this error, if any.

fn description(&self) -> &str

👎Deprecated since 1.42.0:

use the Display impl or to_string()

fn cause(&self) -> Option<&dyn Error>

👎Deprecated since 1.33.0:

replaced by Error::source, which can support downcasting

fn provide<'a>(&'a self, request: &mut Request<'a>)

🔬This is a nightly-only experimental API. (error_generic_member_access)

Provides type-based access to context intended for error reports.

impl From<LlmClientError> for LlmStreamError

fn from(source: LlmClientError) -> Self

Converts to this type from the input type.