Package: llama.cpp (5760+dfsg-2)

[ Source: llama.cpp  ]

LLM inference in C/C++

The main goal of llama.cpp is to enable LLM inference with minimal setup and state-of-the-art performance on a wide range of hardware - locally and in the cloud.

 * Plain C/C++ implementation without any dependencies
 * Apple silicon is a first-class citizen - optimized via ARM NEON, Accelerate
   and Metal frameworks
 * AVX, AVX2, AVX512 and AMX support for x86 architectures
 * 1.5-bit, 2-bit, 3-bit, 4-bit, 5-bit, 6-bit, and 8-bit integer quantization
   for faster inference and reduced memory use
 * Custom CUDA kernels for running LLMs on NVIDIA GPUs (support for AMD GPUs
   via HIP and Moore Threads MTT GPUs via MUSA)
 * Vulkan and SYCL backend support
 * CPU+GPU hybrid inference to partially accelerate models larger than the
   total VRAM capacity

The compute functionality is provided by ggml. By default, ggml's CPU backend is installed, but there are many other backends for CPUs and GPUs.

Other Packages Related to llama.cpp

• dep: libc6 (>= 2.38) [not alpha, loong64]

  GNU C Library: Shared libraries
  also a virtual package provided by libc6-udeb

  dep: libc6 (>= 2.41) [loong64]

• dep: libc6.1 (>= 2.38) [alpha]

  GNU C Library: Shared libraries
  also a virtual package provided by libc6.1-udeb

• dep: libcurl4t64 (>= 7.16.2)

  easy-to-use client-side URL transfer library (OpenSSL flavour)

• dep: libgcc-s1 (>= 3.0) [amd64, arm64, loong64, mips64el, s390x]

  GCC support library

  dep: libgcc-s1 (>= 3.4) [alpha]

  dep: libgcc-s1 (>= 3.4.4) [ppc64, ppc64el]

  dep: libgcc-s1 (>= 4.3) [riscv64, sparc64]

• dep: libggml-cpu (<< 0.0~git20250622)

  Tensor library for machine learning - CPU backend

  or libggml-backend (<< 0.0~git2025062)

  Package not available

  dep: libggml-cpu (>= 0.0~git20250621.b2a092a)

  or libggml-backend (>= 0.0~git20250621.b2a092a)

• dep: libstdc++6 (>= 14)

  GNU Standard C++ Library v3

• dep: python3

  interactive high-level object-oriented language (default python3 version)

Download llama.cpp

Download for all available architectures

Architecture	Package Size	Installed Size	Files
alpha (unofficial port)	7,593.4 kB	66,900.0 kB	[list of files]
amd64	7,879.3 kB	59,849.0 kB	[list of files]
arm64	7,113.6 kB	57,969.0 kB	[list of files]
loong64 (unofficial port)	6,871.8 kB	57,577.0 kB	[list of files]
mips64el	7,004.0 kB	74,025.0 kB	[list of files]
ppc64 (unofficial port)	7,289.8 kB	73,009.0 kB	[list of files]
ppc64el	7,335.5 kB	68,721.0 kB	[list of files]
riscv64	8,007.3 kB	51,361.0 kB	[list of files]
s390x	7,212.6 kB	61,580.0 kB	[list of files]
sparc64 (unofficial port)	6,736.1 kB	101,578.0 kB	[list of files]