Vllm.Config.EPLBConfig (VLLM v0.3.0)

Copy Markdown View Source

Configuration for Expert Parallel Load Balancing (EP).

Summary

Types

t()

@opaque t()

Functions

_validate_eplb_config(ref, opts \\ [])

@spec _validate_eplb_config(
  SnakeBridge.Ref.t(),
  keyword()
) :: {:ok, term()} | {:error, Snakepit.Error.t()}

Python method EPLBConfig._validate_eplb_config.

Returns

  • term()

log_balancedness(ref)

@spec log_balancedness(SnakeBridge.Ref.t()) ::
  {:ok, term()} | {:error, Snakepit.Error.t()}

log_balancedness_interval(ref)

@spec log_balancedness_interval(SnakeBridge.Ref.t()) ::
  {:ok, term()} | {:error, Snakepit.Error.t()}

new(dataclass_self__, args, kwargs, opts \\ [])

@spec new(term(), term(), term(), keyword()) ::
  {:ok, SnakeBridge.Ref.t()} | {:error, Snakepit.Error.t()}

Constructs EPLBConfig.

Parameters

  • dataclass_self__ (term())
  • args (term())
  • kwargs (term())

num_redundant_experts(ref)

@spec num_redundant_experts(SnakeBridge.Ref.t()) ::
  {:ok, term()} | {:error, Snakepit.Error.t()}

policy(ref)

@spec policy(SnakeBridge.Ref.t()) :: {:ok, term()} | {:error, Snakepit.Error.t()}

step_interval(ref)

@spec step_interval(SnakeBridge.Ref.t()) ::
  {:ok, term()} | {:error, Snakepit.Error.t()}

use_async(ref)

@spec use_async(SnakeBridge.Ref.t()) :: {:ok, term()} | {:error, Snakepit.Error.t()}

window_size(ref)

@spec window_size(SnakeBridge.Ref.t()) :: {:ok, term()} | {:error, Snakepit.Error.t()}