Struct Request 

pub struct Request {
    pub id: RequestId,
    pub input_ids: Vec<u32>,
    pub max_tokens: usize,
    pub generated_tokens: Vec<u32>,
    pub priority: Priority,
    pub state: RequestState,
    pub created_at: Instant,
    pub seq_id: Option<usize>,
}

Inference request

Fields

id: RequestId

Unique identifier

input_ids: Vec<u32>

Input token IDs

max_tokens: usize

Maximum tokens to generate

generated_tokens: Vec<u32>

Tokens generated so far

priority: Priority

Request priority

state: RequestState

Current state

created_at: Instant

Creation time

seq_id: Option<usize>

Sequence ID in block manager (if running)

Implementations

impl Request

pub fn new(input_ids: Vec<u32>, max_tokens: usize) -> Self

pub fn with_priority(self, priority: Priority) -> Self

pub fn seq_len(&self) -> usize

Total sequence length (input + generated)

pub fn is_complete(&self) -> bool

Check if generation is complete

Trait Implementations

impl Clone for Request

fn clone(&self) -> Request

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Request

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

Formats the value using the given formatter.