mcpp index

ggml-org.llamacpp

import

C++23 module package for llama.cpp b10069 (import llamacpp;)

import llamacpp;

Module name read from the upstream `export module` declaration in src/llamacpp.cppm.

mcpp add ggml-org.llamacpp@b10069

Usage

From this repository's own test project — built and run by CI, not written for the website.

import std;

#if defined(__APPLE__) && defined(__aarch64__)
import llamacpp;

int main() {
    llama_backend_init();
    const bool metal_available = llama_supports_gpu_offload();
    llama_backend_free();

    if (!metal_available) {
        std::cerr << "Metal backend is not available\n";
        return 1;
    }

    std::cout << "LLAMACPP_METAL_INDEX_TEST=PASS\n";
    return 0;
}
#else
int main() {
    return 0;
}
// …

tests/examples/llamacpp-metal/tests/backend.cpp · project tests/examples/llamacpp-metal

Build · upstream manifest

modules
llamacpp
targets
llama (lib)
language
c++23
manifest version
b10069
source
https://raw.githubusercontent.com/mcpplibs/llama.cpp-m/b10069/mcpp.toml

Versions

VersionPlatforms Mirrorssha256
b10069 latest LinuxWindowsmacOS source be0b0deeb311…