Table of Contents

Namespace LMKit.Global

Classes

Configuration

A static class providing global configuration settings for the LM-Kit runtime.

EmbeddingPoolOptions

Shape and allocation policy of the resident embedding pool: the single pinned context each embedding-capable model decodes its embedding and reranking requests on.

Runtime

Manages the lifecycle of the LM-Kit runtime environment.

Responsibilities include:

  • Native backend initialization and configuration.
  • GPU capability checks.
  • Model lifecycle tracking for licensing and runtime behavior.
  • Centralized logging configuration for diagnostics and customer support.
SharedSlotPoolOptions

Shape and admission policy of the resident slot pool used when InferenceScheduling is SharedSlotPool.

Enums

InferenceScheduling

How the runtime allocates inference state and schedules requests onto it. Selected once for the process; the two strategies are never mixed.

Runtime.BackendType

Specifies the type of computational backend to use.

Runtime.LMKitLogLevel

Defines the log levels available for use within the LM-Kit framework. Each level indicates the severity and verbosity of the log messages.

SharedSlotPoolOptions.AllocationTiming

When the pool's memory is claimed.

SharedSlotPoolOptions.SaturationPolicy

What a request does when every slot is already busy.

Share