ggml

(★ 15,212)

머신러닝을 위한 텐서 라이브러리

파일 탐색기

최종 버전 다운로드 (.zip)

- - build-cpu.yml
  - build-self-hosted.yml
  - make-release.yml
  - release.yml
- pull_request_template.md
- - SYSTEM.md
- SYSTEM.md
- run.sh
- common.cmake
- FindNCCL.cmake
- ggml-config.cmake.in
- GitVars.cmake
- gguf.md
- - CMakeLists.txt
  - convert-cerebras-to-ggml.py
  - convert-ckpt-to-ggml.py
  - convert-h5-to-ggml.py
  - download-ggml-model.sh
  - download-model.sh
  - main-alloc.cpp
  - main-backend.cpp
  - main-batched.cpp
  - main-ctx.cpp
  - main-sched.cpp
  - quantize.cpp
  - README.md
- - CMakeLists.txt
  - convert-h5-to-ggml.py
  - download-ggml-model.sh
  - download-model.sh
  - main.cpp
  - quantize.cpp
  - README.md
- - CMakeLists.txt
  - convert.py
  - main.cpp
  - README.md
- - .gitignore
  - CMakeLists.txt
  - mnist-common.cpp
  - mnist-common.h
  - mnist-eval.cpp
  - mnist-train-cnn.py
  - mnist-train-fc.py
  - mnist-train.cpp
  - README.md
  - server.py
- - CMakeLists.txt
  - perf-metal.cpp
- - dolly-v2.txt
  - gpt-2-chinese.txt
  - gpt-2.txt
  - gpt-j.txt
  - gpt-neox-japanese.txt
  - gpt-neox.txt
  - polyglot-ko.txt
  - replit.txt
  - starcoder.txt
  - test-cases.txt
  - tokenize_huggingface.py
  - whisper.txt
- - api.h
  - example_add_quant.py
  - example_test_all_quants.py
  - README.md
  - regenerate.py
  - stubs.py
  - test_tensor.py
- - CMakeLists.txt
  - convert-pth-to-ggml.py
  - example.jpg
  - README.md
  - sam.cpp
- - CMakeLists.txt
  - README.md
  - simple-backend.cpp
  - simple-ctx.cpp
- - .gitignore
  - build-install.sh
  - CMakeLists.txt
  - README.md
  - test-cmake.cpp
- - CMakeLists.txt
  - convert-yolov3-tiny.py
  - README.md
  - yolo-image.cpp
  - yolo-image.h
  - yolov3-tiny.cpp
- CMakeLists.txt
- common-ggml.cpp
- common-ggml.h
- common.cpp
- common.h
- stb_image.h
- stb_image_write.h
- ggml-alloc.h
- ggml-backend.h
- ggml-blas.h
- ggml-cann.h
- ggml-cpp.h
- ggml-cpu.h
- ggml-cuda.h
- ggml-et.h
- ggml-hexagon.h
- ggml-metal.h
- ggml-opencl.h
- ggml-openvino.h
- ggml-opt.h
- ggml-rpc.h
- ggml-sycl.h
- ggml-virtgpu.h
- ggml-vulkan.h
- ggml-webgpu.h
- ggml-zdnn.h
- ggml-zendnn.h
- ggml.h
- gguf.h
- gen-authors.sh
- make-release-checks.sh
- make-release-desc.sh
- release.sh
- sync-llama-am.sh
- sync-llama.last
- sync-llama.sh
- sync-whisper-am.sh
- sync-whisper.last
- sync-whisper.sh
- - CMakeLists.txt
  - ggml-blas.cpp
- - acl_tensor.cpp
  - acl_tensor.h
  - aclnn_ops.cpp
  - aclnn_ops.h
  - CMakeLists.txt
  - common.h
  - ggml-cann.cpp
- - arch-fallback.h
  - binary-ops.cpp
  - binary-ops.h
  - CMakeLists.txt
  - common.h
  - ggml-cpu-impl.h
  - ggml-cpu.c
  - ggml-cpu.cpp
  - hbm.cpp
  - hbm.h
  - ops.cpp
  - ops.h
  - quants.c
  - quants.h
  - repack.cpp
  - repack.h
  - simd-gemm.h
  - simd-mappings.h
  - traits.cpp
  - traits.h
  - unary-ops.cpp
  - unary-ops.h
  - vec.cpp
  - vec.h
- - acc.cu
  - acc.cuh
  - add-id.cu
  - add-id.cuh
  - allreduce.cu
  - allreduce.cuh
  - arange.cu
  - arange.cuh
  - argmax.cu
  - argmax.cuh
  - argsort.cu
  - argsort.cuh
  - binbcast.cu
  - binbcast.cuh
  - clamp.cu
  - clamp.cuh
  - CMakeLists.txt
  - col2im-1d.cu
  - col2im-1d.cuh
  - common.cuh
  - concat.cu
  - concat.cuh
  - conv-transpose-1d.cu
  - conv-transpose-1d.cuh
  - conv2d-dw.cu
  - conv2d-dw.cuh
  - conv2d-transpose.cu
  - conv2d-transpose.cuh
  - conv2d.cu
  - conv2d.cuh
  - convert.cu
  - convert.cuh
  - count-equal.cu
  - count-equal.cuh
  - cp-async.cuh
  - cpy-utils.cuh
  - cpy.cu
  - cpy.cuh
  - cross-entropy-loss.cu
  - cross-entropy-loss.cuh
  - cumsum.cu
  - cumsum.cuh
  - dequantize.cuh
  - diag.cu
  - diag.cuh
  - diagmask.cu
  - diagmask.cuh
  - dsv4-hc.cu
  - dsv4-hc.cuh
  - fattn-common.cuh
  - fattn-mma-f16.cuh
  - fattn-tile.cu
  - fattn-tile.cuh
  - fattn-vec.cuh
  - fattn.cu
  - fattn.cuh
  - fill.cu
  - fill.cuh
  - fwht.cu
  - fwht.cuh
  - gated_delta_net.cu
  - gated_delta_net.cuh
  - getrows.cu
  - getrows.cuh
  - ggml-cuda.cu
  - gla.cu
  - gla.cuh
  - im2col.cu
  - im2col.cuh
  - lightning-indexer.cu
  - lightning-indexer.cuh
  - mean.cu
  - mean.cuh
  - mma.cuh
  - mmf.cu
  - mmf.cuh
  - mmid.cu
  - mmid.cuh
  - mmq-config-ampere.cuh
  - mmq-config-blackwell.cuh
  - mmq-config-cdna.cuh
  - mmq-config-pascal.cuh
  - mmq-config-rdna2.cuh
  - mmq-config-rdna3-5.cuh
  - mmq-config-rdna3.cuh
  - mmq-config-rdna4.cuh
  - mmq-load-tiles.cuh
  - mmq-vec-dot.cuh
  - mmq.cu
  - mmq.cuh
  - mmvf.cu
  - mmvf.cuh
  - mmvq.cu
  - mmvq.cuh
  - norm.cu
  - norm.cuh
  - opt-step-adamw.cu
  - opt-step-adamw.cuh
  - opt-step-sgd.cu
  - opt-step-sgd.cuh
  - out-prod.cu
  - out-prod.cuh
  - pad.cu
  - pad.cuh
  - pad_reflect_1d.cu
  - pad_reflect_1d.cuh
  - pool2d.cu
  - pool2d.cuh
  - quantize.cu
  - quantize.cuh
  - reduce_rows.cuh
  - roll.cu
  - roll.cuh
  - rope.cu
  - rope.cuh
  - scale.cu
  - scale.cuh
  - set-rows.cu
  - set-rows.cuh
  - set.cu
  - set.cuh
  - snake.cu
  - snake.cuh
  - softcap.cu
  - softcap.cuh
  - softmax.cu
  - softmax.cuh
  - solve_tri.cu
  - solve_tri.cuh
  - ssm-conv.cu
  - ssm-conv.cuh
  - ssm-scan.cu
  - ssm-scan.cuh
  - sum.cu
  - sum.cuh
  - sumrows.cu
  - sumrows.cuh
  - top-k.cu
  - top-k.cuh
  - topk-moe.cu
  - topk-moe.cuh
  - tri.cu
  - tri.cuh
  - tsembd.cu
  - tsembd.cuh
  - unary.cu
  - unary.cuh
  - upscale.cu
  - upscale.cuh
  - vecdotq.cuh
  - wkv.cu
  - wkv.cuh
- - CMakeLists.txt
  - ggml-et-common.h
  - ggml-et-cpu-compare.cpp
  - ggml-et-cpu-compare.h
  - ggml-et-kernels.cpp
  - ggml-et-kernels.h
  - ggml-et-memops.cpp
  - ggml-et-memops.h
  - ggml-et-ops.cpp
  - ggml-et-ops.h
  - ggml-et-uberkernel-common.h
  - ggml-et.cpp
- - CMakeLists.txt
  - ggml-hexagon.cpp
  - htp-drv.cpp
  - htp-drv.h
  - htp-opnode.h
  - libdl.h
  - libggml-htp.inf
- - CMakeLists.txt
- - CMakeLists.txt
  - ggml-metal-common.cpp
  - ggml-metal-common.h
  - ggml-metal-context.h
  - ggml-metal-context.m
  - ggml-metal-device.cpp
  - ggml-metal-device.h
  - ggml-metal-device.m
  - ggml-metal-impl.h
  - ggml-metal-ops.cpp
  - ggml-metal-ops.h
  - ggml-metal.cpp
  - ggml-metal.metal
- - CMakeLists.txt
  - mudnn.cu
  - mudnn.cuh
- - cl-program-cache.cpp
  - cl-program-cache.h
  - CMakeLists.txt
  - fa_tune.h
  - ggml-opencl.cpp
  - libdl.h
- - .clang-format
  - CMakeLists.txt
  - ggml-decoder.cpp
  - ggml-decoder.h
  - ggml-openvino-extra.cpp
  - ggml-openvino-extra.h
  - ggml-openvino.cpp
  - ggml-quants.cpp
  - ggml-quants.h
  - model-cache.cpp
  - model-cache.h
  - utils.cpp
  - utils.h
- - CMakeLists.txt
  - ggml-rpc.cpp
  - transport.cpp
  - transport.h
- - add-id.cpp
  - add-id.hpp
  - backend.hpp
  - binbcast.cpp
  - binbcast.hpp
  - CMakeLists.txt
  - col2im-1d.cpp
  - col2im-1d.hpp
  - common.cpp
  - common.hpp
  - concat.cpp
  - concat.hpp
  - conv.cpp
  - conv.hpp
  - conv2d-dw.cpp
  - conv2d-dw.hpp
  - conv2d-transpose.cpp
  - conv2d-transpose.hpp
  - conv2d.cpp
  - conv2d.hpp
  - conv3d.cpp
  - conv3d.hpp
  - convert.cpp
  - convert.hpp
  - count-equal.cpp
  - count-equal.hpp
  - cpy.cpp
  - cpy.hpp
  - cross_entropy_loss.cpp
  - cross_entropy_loss.hpp
  - cumsum.cpp
  - cumsum.hpp
  - dequantize.hpp
  - diag.cpp
  - diag.hpp
  - dmmv.cpp
  - dmmv.hpp
  - dsv4-hc.cpp
  - dsv4-hc.hpp
  - element_wise.cpp
  - element_wise.hpp
  - esimd.hpp
  - fattn-buffers.cpp
  - fattn-buffers.hpp
  - fattn-common.hpp
  - fattn-mkl.cpp
  - fattn-onednn.cpp
  - fattn-onednn.hpp
  - fattn-tile.cpp
  - fattn-tile.hpp
  - fattn-vec.hpp
  - fattn.cpp
  - fattn.hpp
  - fill.cpp
  - fill.hpp
  - fusion.cpp
  - fusion.hpp
  - fwht.cpp
  - fwht.hpp
  - gated_delta_net.cpp
  - gated_delta_net.hpp
  - gemm.hpp
  - getrows.cpp
  - getrows.hpp
  - ggml-sycl.cpp
  - gla.cpp
  - gla.hpp
  - im2col.cpp
  - im2col.hpp
  - lightning-indexer.cpp
  - lightning-indexer.hpp
  - mmq.cpp
  - mmq.hpp
  - mmvq.cpp
  - mmvq.hpp
  - norm.cpp
  - norm.hpp
  - opt-step.cpp
  - opt-step.hpp
  - outprod.cpp
  - outprod.hpp
  - pad.cpp
  - pad.hpp
  - pad_reflect_1d.cpp
  - pad_reflect_1d.hpp
  - pool.cpp
  - pool.hpp
  - presets.hpp
  - quantize.hpp
  - quants.hpp
  - repeat_back.cpp
  - repeat_back.hpp
  - roll.cpp
  - roll.hpp
  - rope.cpp
  - rope.hpp
  - set.cpp
  - set.hpp
  - set_rows.cpp
  - set_rows.hpp
  - softmax.cpp
  - softmax.hpp
  - solve_tri.cpp
  - solve_tri.hpp
  - ssm_conv.cpp
  - ssm_conv.hpp
  - ssm_scan.cpp
  - ssm_scan.hpp
  - sycl_hw.cpp
  - sycl_hw.hpp
  - topk-moe.cpp
  - topk-moe.hpp
  - tsembd.cpp
  - tsembd.hpp
  - type.hpp
  - upscale.cpp
  - upscale.hpp
  - vecdotq.hpp
  - wkv.cpp
  - wkv.hpp
- - apir_cs_ggml-rpc-front.cpp
  - CMakeLists.txt
  - ggml-backend-buffer-type.cpp
  - ggml-backend-buffer.cpp
  - ggml-backend-device.cpp
  - ggml-backend-reg.cpp
  - ggml-backend.cpp
  - ggml-remoting.h
  - ggmlremoting_functions.yaml
  - regenerate_remoting.py
  - virtgpu-apir.h
  - virtgpu-forward-backend.cpp
  - virtgpu-forward-buffer-type.cpp
  - virtgpu-forward-buffer.cpp
  - virtgpu-forward-device.cpp
  - virtgpu-forward-impl.h
  - virtgpu-forward.gen.h
  - virtgpu-shm.cpp
  - virtgpu-shm.h
  - virtgpu-utils.cpp
  - virtgpu-utils.h
  - virtgpu.cpp
  - virtgpu.h
- - CMakeLists.txt
  - ggml-vulkan.cpp
- - CMakeLists.txt
  - ggml-webgpu-shader-lib.hpp
  - ggml-webgpu.cpp
  - pre_wgsl.hpp
- - .gitignore
  - CMakeLists.txt
  - common.hpp
  - ggml-zdnn.cpp
  - mmf.cpp
  - mmf.hpp
  - utils.cpp
  - utils.hpp
- - CMakeLists.txt
  - ggml-zendnn.cpp
- CMakeLists.txt
- ggml-alloc.c
- ggml-backend-dl.cpp
- ggml-backend-dl.h
- ggml-backend-impl.h
- ggml-backend-meta.cpp
- ggml-backend-reg.cpp
- ggml-backend.cpp
- ggml-common.h
- ggml-feats.h
- ggml-impl.h
- ggml-opt.cpp
- ggml-quants.c
- ggml-quants.h
- ggml-threading.cpp
- ggml-threading.h
- ggml.c
- ggml.cpp
- gguf.cpp
- CMakeLists.txt
- test-arange.cpp
- test-backend-ops.cpp
- test-cont.c
- test-conv-transpose-1d.cpp
- test-conv-transpose.c
- test-conv1d-dw-c1.cpp
- test-conv1d-dw-c2.cpp
- test-conv1d.cpp
- test-conv2d-dw.cpp
- test-conv2d.cpp
- test-customop.c
- test-dup.c
- test-interpolate.cpp
- test-opt.cpp
- test-pad-reflect-1d.cpp
- test-pool.c
- test-quantize-fns.cpp
- test-quantize-perf.cpp
- test-rel-pos.c
- test-roll.cpp
- test-timestep_embedding.cpp
.editorconfig
.gitignore
.gitmodules
AUTHORS
CMakeLists.txt
CONTRIBUTING.md
ggml.pc.in
LICENSE
README.md
requirements.txt

# 설치 가이드

1. 코드 내려받기

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

깃허브에서 프로젝트 코드 전체를 내 컴퓨터로 내려받습니다.

cd ggml

방금 내려받은 프로젝트 폴더 안으로 이동합니다.

2. CMake

보통 추천

사전 준비물

Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
CMake 빌드 설정을 생성하는 도구입니다.
C/C++ 컴파일러 Windows는 Visual Studio(Community 무료), macOS는 Xcode Command Line Tools, Linux는 gcc/g++ 패키지가 필요합니다.

cmake ..

소스코드를 분석해 빌드 설정 파일을 생성합니다 (build 폴더 안에서 실행해야 함).

cmake --build . --config Release -j 8

이 명령어를 터미널에 그대로 입력해 실행하세요.

✅ build 폴더 안에 실행 파일이 생성됐는지 확인하고, 직접 실행해보세요 (예: ./build/앱이름).

이 레포의 README에 적힌 실제 명령어를 그대로 가져왔습니다.

3. Python

쉬움

사전 준비물

Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
Python 3 설치 시 'Add Python to PATH' 옵션을 꼭 체크하세요 (Windows).

pip install -r requirements.txt

requirements.txt 등에 명시된 파이썬 라이브러리를 설치합니다.

python <실행할 파일명>.py # README에서 정확한 실행 파일명을 확인하세요

파이썬 스크립트(또는 모듈)를 실행합니다.

✅ 에러 메시지 없이 실행되고 터미널에 안내 문구가 출력되면 정상입니다.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

ggml

파일 탐색기

# 설치 가이드

2. CMake

3. Python

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

ggml

파일 탐색기

# 설치 가이드

2. CMake

3. Python

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README