liblloyal 1.0.0
Branched Inference for llama.cpp
Loading...
Searching...
No Matches
lloyal::ModelKey Struct Reference

Model cache key combining file path and GPU configuration. More...

#include <lloyal/model_registry.hpp>

Public Member Functions

bool operator== (const ModelKey &o) const
 

Public Attributes

std::string canonPath
 Normalized file path (file:// prefix removed)
 
int n_gpu_layers
 Number of layers offloaded to GPU (-1 = all)
 
int load_mode
 llama_load_mode the model was loaded with
 

Detailed Description

Model cache key combining file path and GPU configuration.

Used as key in ModelRegistry cache to uniquely identify model instances. Different GPU configurations of the same model are cached separately.

Definition at line 36 of file model_registry.hpp.

Member Function Documentation

◆ operator==()

bool lloyal::ModelKey::operator== ( const ModelKey &  o) const
inline

Member Data Documentation

◆ canonPath

std::string lloyal::ModelKey::canonPath

Normalized file path (file:// prefix removed)

Examples
/home/runner/work/liblloyal/liblloyal/include/lloyal/model_registry.hpp.

Definition at line 37 of file model_registry.hpp.

◆ load_mode

int lloyal::ModelKey::load_mode

llama_load_mode the model was loaded with

Examples
/home/runner/work/liblloyal/liblloyal/include/lloyal/model_registry.hpp.

Definition at line 39 of file model_registry.hpp.

◆ n_gpu_layers

int lloyal::ModelKey::n_gpu_layers

Number of layers offloaded to GPU (-1 = all)

Examples
/home/runner/work/liblloyal/liblloyal/include/lloyal/model_registry.hpp.

Definition at line 38 of file model_registry.hpp.


The documentation for this struct was generated from the following file: