Struct CompletedRequest 

pub struct CompletedRequest {
    pub id: RequestId,
    pub prompt_tokens: Vec<u32>,
    pub generated_tokens: Vec<u32>,
    pub state: RequestState,
    pub processing_time_ms: u64,
    pub waiting_time_ms: u64,
    pub prefill_time_ms: u64,
    pub decode_time_ms: u64,
    pub decode_steps: usize,
    pub tokens_per_second: f64,
    pub error: Option<String>,
    pub finish_reason: FinishReason,
}

Result of a completed request

Fields

id: RequestId

Request ID

prompt_tokens: Vec<u32>

Original prompt tokens

generated_tokens: Vec<u32>

Generated tokens

state: RequestState

Final state

processing_time_ms: u64

Total processing time

waiting_time_ms: u64

Time spent waiting

prefill_time_ms: u64

Prefill time

decode_time_ms: u64

Decode time

decode_steps: usize

Number of decode steps

tokens_per_second: f64

Tokens per second during decode

error: Option<String>

Error message if failed

finish_reason: FinishReason

Finish reason

Implementations

impl CompletedRequest

pub fn success(running: &RunningRequest, prefill_time_ms: u64) -> Self

Create a successful completion

pub fn failure(running: &RunningRequest, error: impl Into<String>) -> Self

Create a failed completion

pub fn cancelled(running: &RunningRequest) -> Self

Create a cancelled completion

pub fn total_tokens(&self) -> usize

Get total token count

Trait Implementations

impl Clone for CompletedRequest

fn clone(&self) -> CompletedRequest

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for CompletedRequest

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

Formats the value using the given formatter.

impl From<CompletedRequest> for GenerationResult

fn from(completed: CompletedRequest) -> Self

Converts to this type from the input type.