Tensor library for machine learning https://github.com/ggml-org
  • C++ 54.5%
  • C 28.6%
  • Cuda 9%
  • Metal 2.6%
  • GLSL 1.5%
  • Other 3.8%
Find a file
Repository files (latest commit first)
Filename Latest commit message Latest commit date
Daniel Bevenius d9fe7bff3d
examples : remove explict version from test-cmake (#1692)
This commit removes the ggml version test-cmake and instead uses the
version that is supplied by CMAKE_PREFIX_PATH.

The motivation for this is to avoid issues when bumping major versions.

Refs: https://github.com/ggml-org/ggml/actions/runs/43385979342/job/131763861057?pr=1690
2026-11-16 10:35:03 +01:00
.github ci : upload API/ABI reports as artifacts (#1666) 2026-10-20 14:20:49 +03:00
.pi/gg pi : turn .pi/gg into a system prompt addendum 2026-10-20 13:45:49 +03:00
ci ggml : fix conv2d_dw SVE path (#1380) 2025-11-04 20:40:52 +02:00
cmake CUDA: Add OCP FP8 support (llama/28413) 2026-11-12 21:02:21 +02:00
docs docs : sync ggml_type enum in gguf.md (#1685) 2026-11-08 15:56:21 +02:00
examples examples : remove explict version from test-cmake (#1692) 2026-11-16 10:35:03 +01:00
include ggml : prune (llama/29801) 2026-11-12 21:02:22 +02:00
scripts sync : llama.cpp 2026-11-13 09:06:53 +02:00
src fix(hexagon): make host build work with clang-cl on Windows ARM64 (llama/32011) 2026-11-13 09:06:47 +02:00
tests ggml : prune (llama/29801) 2026-11-12 21:02:22 +02:00
.editorconfig gguf : add file format specification (#302) 2023-11-01 19:01:49 +02:00
.gitignore pi : turn .pi/gg into a system prompt addendum 2026-10-20 13:45:49 +03:00
.gitmodules git : remove kompute submodule (#1300) 2025-07-12 16:12:49 +03:00
AUTHORS ggml : bump version to 0.29.0 (#1680) 2026-11-04 11:27:44 +02:00
CMakeLists.txt ggml : bump version to 1.0.0 (#1690) 2026-11-13 10:13:55 +02:00
CONTRIBUTING.md contrib : recommend PRs to llama.cpp (#1312) 2025-07-25 07:05:38 +03:00
ggml.pc.in pkg-config: include the new GGML_VERSION as a version (#1348) 2025-09-25 18:59:38 +02:00
LICENSE docs : Minor cleanups (llama/19252) 2026-02-07 10:37:38 +02:00
README.md readme : update following llama.cpp style (#1599) 2026-08-21 15:39:22 +03:00
requirements.txt requirements : fix numpy for python 3.14 (#1691) 2026-11-13 10:13:40 +02:00

ggml

ggml logo

Tensor library for machine learning

License: MIT Release CI

Quick start

Build from source:

git clone https://github.com/ggml-org/ggml
cd ggml

mkdir build && cd build
cmake ..
cmake --build . --config Release -j 8

For a minimal, fully commented example (matrix multiplication), see examples/simple.

Description

The main goal of ggml is to be a simple, portable, and efficient tensor library for machine learning with minimal setup.

  • Plain C/C++ implementation without any dependencies
  • Cross-platform - x86, ARM, RISC-V, LoongArch, PowerPC, s390x, and WebAssembly
  • SIMD-optimized kernels for x86, ARM, and RISC-V
  • Broad backend support - CPU, GPU, NPU, and browser
  • 2- to 8-bit integer quantization, plus MXFP4 and NVFP4 microscaling formats
  • Zero memory allocations during runtime

Documentation

Contributing

  • For changes to the core ggml library (including to the CMake build system), please open a PR in llama.cpp - doing so will make your PR more visible, better tested, and more likely to be reviewed