Skip to content

Commit

Permalink
Merge pull request #34 from l3utterfly/master
Browse files Browse the repository at this point in the history
merge upstream
  • Loading branch information
l3utterfly authored Aug 28, 2024
2 parents c4934bc + 23e298c commit 3437c58
Show file tree
Hide file tree
Showing 128 changed files with 12,969 additions and 9,038 deletions.
44 changes: 44 additions & 0 deletions .devops/llama-cli-cann.Dockerfile
Original file line number Diff line number Diff line change
@@ -0,0 +1,44 @@
ARG ASCEND_VERSION=8.0.rc2.alpha003-910b-openeuler22.03-py3.8

FROM cosdt/cann:$ASCEND_VERSION AS build

WORKDIR /app

COPY . .

RUN yum install -y gcc g++ cmake make
ENV ASCEND_TOOLKIT_HOME=/usr/local/Ascend/ascend-toolkit/latest
ENV LIBRARY_PATH=${ASCEND_TOOLKIT_HOME}/lib64:$LIBRARY_PATH
ENV LD_LIBRARY_PATH=${ASCEND_TOOLKIT_HOME}/lib64:${ASCEND_TOOLKIT_HOME}/lib64/plugin/opskernel:${ASCEND_TOOLKIT_HOME}/lib64/plugin/nnengine:${ASCEND_TOOLKIT_HOME}/opp/built-in/op_impl/ai_core/tbe/op_tiling:${LD_LIBRARY_PATH}
ENV PYTHONPATH=${ASCEND_TOOLKIT_HOME}/python/site-packages:${ASCEND_TOOLKIT_HOME}/opp/built-in/op_impl/ai_core/tbe:${PYTHONPATH}
ENV PATH=${ASCEND_TOOLKIT_HOME}/bin:${ASCEND_TOOLKIT_HOME}/compiler/ccec_compiler/bin:${PATH}
ENV ASCEND_AICPU_PATH=${ASCEND_TOOLKIT_HOME}
ENV ASCEND_OPP_PATH=${ASCEND_TOOLKIT_HOME}/opp
ENV TOOLCHAIN_HOME=${ASCEND_TOOLKIT_HOME}/toolkit
ENV ASCEND_HOME_PATH=${ASCEND_TOOLKIT_HOME}

# find libascend_hal.so, because the drive hasn`t been mounted.
ENV LD_LIBRARY_PATH=${ASCEND_TOOLKIT_HOME}/runtime/lib64/stub:$LD_LIBRARY_PATH

RUN echo "Building with static libs" && \
source /usr/local/Ascend/ascend-toolkit/set_env.sh --force && \
cmake -B build -DGGML_CANN=ON -DBUILD_SHARED_LIBS=OFF && \
cmake --build build --config Release --target llama-cli

# TODO: use image with NNRT
FROM cosdt/cann:$ASCEND_VERSION AS runtime
COPY --from=build /app/build/bin/llama-cli /llama-cli

ENV LC_ALL=C.utf8

ENV ASCEND_TOOLKIT_HOME=/usr/local/Ascend/ascend-toolkit/latest
ENV LIBRARY_PATH=${ASCEND_TOOLKIT_HOME}/lib64:$LIBRARY_PATH
ENV LD_LIBRARY_PATH=${ASCEND_TOOLKIT_HOME}/lib64:${ASCEND_TOOLKIT_HOME}/lib64/plugin/opskernel:${ASCEND_TOOLKIT_HOME}/lib64/plugin/nnengine:${ASCEND_TOOLKIT_HOME}/opp/built-in/op_impl/ai_core/tbe/op_tiling:${LD_LIBRARY_PATH}
ENV PYTHONPATH=${ASCEND_TOOLKIT_HOME}/python/site-packages:${ASCEND_TOOLKIT_HOME}/opp/built-in/op_impl/ai_core/tbe:${PYTHONPATH}
ENV PATH=${ASCEND_TOOLKIT_HOME}/bin:${ASCEND_TOOLKIT_HOME}/compiler/ccec_compiler/bin:${PATH}
ENV ASCEND_AICPU_PATH=${ASCEND_TOOLKIT_HOME}
ENV ASCEND_OPP_PATH=${ASCEND_TOOLKIT_HOME}/opp
ENV TOOLCHAIN_HOME=${ASCEND_TOOLKIT_HOME}/toolkit
ENV ASCEND_HOME_PATH=${ASCEND_TOOLKIT_HOME}

ENTRYPOINT ["/llama-cli" ]
2 changes: 2 additions & 0 deletions .devops/llama-server-cuda.Dockerfile
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,8 @@ ENV CUDA_DOCKER_ARCH=${CUDA_DOCKER_ARCH}
ENV GGML_CUDA=1
# Enable cURL
ENV LLAMA_CURL=1
# Must be set to 0.0.0.0 so it can listen to requests from host machine
ENV LLAMA_ARG_HOST=0.0.0.0

RUN make -j$(nproc) llama-server

Expand Down
2 changes: 2 additions & 0 deletions .devops/llama-server-intel.Dockerfile
Original file line number Diff line number Diff line change
Expand Up @@ -26,6 +26,8 @@ RUN apt-get update && \
COPY --from=build /app/build/bin/llama-server /llama-server

ENV LC_ALL=C.utf8
# Must be set to 0.0.0.0 so it can listen to requests from host machine
ENV LLAMA_ARG_HOST=0.0.0.0

HEALTHCHECK CMD [ "curl", "-f", "http://localhost:8080/health" ]

Expand Down
2 changes: 2 additions & 0 deletions .devops/llama-server-rocm.Dockerfile
Original file line number Diff line number Diff line change
Expand Up @@ -39,6 +39,8 @@ ENV GPU_TARGETS=${ROCM_DOCKER_ARCH}
ENV GGML_HIPBLAS=1
ENV CC=/opt/rocm/llvm/bin/clang
ENV CXX=/opt/rocm/llvm/bin/clang++
# Must be set to 0.0.0.0 so it can listen to requests from host machine
ENV LLAMA_ARG_HOST=0.0.0.0

# Enable cURL
ENV LLAMA_CURL=1
Expand Down
2 changes: 2 additions & 0 deletions .devops/llama-server-vulkan.Dockerfile
Original file line number Diff line number Diff line change
Expand Up @@ -23,6 +23,8 @@ RUN cp /app/build/bin/llama-server /llama-server && \
rm -rf /app

ENV LC_ALL=C.utf8
# Must be set to 0.0.0.0 so it can listen to requests from host machine
ENV LLAMA_ARG_HOST=0.0.0.0

HEALTHCHECK CMD [ "curl", "-f", "http://localhost:8080/health" ]

Expand Down
2 changes: 2 additions & 0 deletions .devops/llama-server.Dockerfile
Original file line number Diff line number Diff line change
Expand Up @@ -21,6 +21,8 @@ RUN apt-get update && \
COPY --from=build /app/llama-server /llama-server

ENV LC_ALL=C.utf8
# Must be set to 0.0.0.0 so it can listen to requests from host machine
ENV LLAMA_ARG_HOST=0.0.0.0

HEALTHCHECK CMD [ "curl", "-f", "http://localhost:8080/health" ]

Expand Down
2 changes: 1 addition & 1 deletion .ecrc
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
{
"Exclude": ["^\\.gitmodules$"],
"Exclude": ["^\\.gitmodules$", "stb_image\\.h"],
"Disable": {
"IndentSize": true
}
Expand Down
Original file line number Diff line number Diff line change
@@ -1,3 +1,6 @@
# TODO: there have been some issues with the workflow, so disabling for now
# https://github.com/ggerganov/llama.cpp/issues/7893
#
# Benchmark
name: Benchmark

Expand Down Expand Up @@ -129,6 +132,8 @@ jobs:

- name: Server bench
id: server_bench
env:
HEAD_REF: ${{ github.head_ref || github.ref_name }}
run: |
set -eux

Expand All @@ -137,7 +142,7 @@ jobs:
python bench.py \
--runner-label ${{ env.RUNNER_LABEL }} \
--name ${{ github.job }} \
--branch ${{ github.head_ref || github.ref_name }} \
--branch $HEAD_REF \
--commit ${{ github.event.inputs.sha || github.event.pull_request.head.sha || github.sha }} \
--scenario script.js \
--duration ${{ github.event.inputs.duration || env.DURATION }} \
Expand Down
22 changes: 10 additions & 12 deletions .github/workflows/build.yml
Original file line number Diff line number Diff line change
Expand Up @@ -47,7 +47,7 @@ jobs:
sysctl -a
mkdir build
cd build
cmake -DLLAMA_FATAL_WARNINGS=ON -DGGML_METAL_EMBED_LIBRARY=ON -DLLAMA_CURL=ON -DBUILD_SHARED_LIBS=OFF ..
cmake -DLLAMA_FATAL_WARNINGS=ON -DGGML_METAL_EMBED_LIBRARY=ON -DLLAMA_CURL=ON -DGGML_RPC=ON -DBUILD_SHARED_LIBS=OFF ..
cmake --build . --config Release -j $(sysctl -n hw.logicalcpu)
- name: Test
Expand Down Expand Up @@ -105,7 +105,7 @@ jobs:
sysctl -a
# Metal is disabled due to intermittent failures with Github runners not having a GPU:
# https://github.com/ggerganov/llama.cpp/actions/runs/8635935781/job/23674807267#step:5:2313
cmake -B build -DLLAMA_FATAL_WARNINGS=ON -DGGML_METAL=OFF -DLLAMA_CURL=ON -DBUILD_SHARED_LIBS=OFF
cmake -B build -DLLAMA_FATAL_WARNINGS=ON -DGGML_METAL=OFF -DLLAMA_CURL=ON -DGGML_RPC=ON -DBUILD_SHARED_LIBS=OFF
cmake --build build --config Release -j $(sysctl -n hw.logicalcpu)
- name: Test
Expand Down Expand Up @@ -222,7 +222,7 @@ jobs:
run: |
mkdir build
cd build
cmake .. -DLLAMA_FATAL_WARNINGS=ON -DLLAMA_CURL=ON -DBUILD_SHARED_LIBS=OFF
cmake .. -DLLAMA_FATAL_WARNINGS=ON -DLLAMA_CURL=ON -DGGML_RPC=ON -DBUILD_SHARED_LIBS=OFF
cmake --build . --config Release -j $(nproc)
- name: Test
Expand Down Expand Up @@ -696,22 +696,20 @@ jobs:
strategy:
matrix:
include:
- build: 'rpc-x64'
defines: '-DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_RPC=ON -DBUILD_SHARED_LIBS=ON'
- build: 'noavx-x64'
defines: '-DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_AVX=OFF -DGGML_AVX2=OFF -DGGML_FMA=OFF -DBUILD_SHARED_LIBS=ON'
defines: '-DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_RPC=ON -DGGML_AVX=OFF -DGGML_AVX2=OFF -DGGML_FMA=OFF -DBUILD_SHARED_LIBS=ON'
- build: 'avx2-x64'
defines: '-DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DBUILD_SHARED_LIBS=ON'
defines: '-DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_RPC=ON -DBUILD_SHARED_LIBS=ON'
- build: 'avx-x64'
defines: '-DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_AVX2=OFF -DBUILD_SHARED_LIBS=ON'
defines: '-DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_RPC=ON -DGGML_AVX2=OFF -DBUILD_SHARED_LIBS=ON'
- build: 'avx512-x64'
defines: '-DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_AVX512=ON -DBUILD_SHARED_LIBS=ON'
defines: '-DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_RPC=ON -DGGML_AVX512=ON -DBUILD_SHARED_LIBS=ON'
- build: 'openblas-x64'
defines: '-DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_BLAS=ON -DBUILD_SHARED_LIBS=ON -DGGML_BLAS_VENDOR=OpenBLAS -DBLAS_INCLUDE_DIRS="$env:RUNNER_TEMP/openblas/include" -DBLAS_LIBRARIES="$env:RUNNER_TEMP/openblas/lib/openblas.lib"'
defines: '-DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_RPC=ON -DGGML_BLAS=ON -DBUILD_SHARED_LIBS=ON -DGGML_BLAS_VENDOR=OpenBLAS -DBLAS_INCLUDE_DIRS="$env:RUNNER_TEMP/openblas/include" -DBLAS_LIBRARIES="$env:RUNNER_TEMP/openblas/lib/openblas.lib"'
- build: 'kompute-x64'
defines: '-DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_KOMPUTE=ON -DKOMPUTE_OPT_DISABLE_VULKAN_VERSION_CHECK=ON -DBUILD_SHARED_LIBS=ON'
defines: '-DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_RPC=ON -DGGML_KOMPUTE=ON -DKOMPUTE_OPT_DISABLE_VULKAN_VERSION_CHECK=ON -DBUILD_SHARED_LIBS=ON'
- build: 'vulkan-x64'
defines: '-DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_VULKAN=ON -DBUILD_SHARED_LIBS=ON'
defines: '-DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_RPC=ON -DGGML_VULKAN=ON -DBUILD_SHARED_LIBS=ON'
- build: 'llvm-arm64'
defines: '-G "Ninja Multi-Config" -D CMAKE_TOOLCHAIN_FILE=cmake/arm64-windows-llvm.cmake -DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DBUILD_SHARED_LIBS=ON'
- build: 'msvc-arm64'
Expand Down
6 changes: 2 additions & 4 deletions .github/workflows/python-check-requirements.yml
Original file line number Diff line number Diff line change
Expand Up @@ -6,15 +6,13 @@ on:
- '.github/workflows/python-check-requirements.yml'
- 'scripts/check-requirements.sh'
- 'convert*.py'
- 'requirements.txt'
- 'requirements/*.txt'
- '**/requirements*.txt'
pull_request:
paths:
- '.github/workflows/python-check-requirements.yml'
- 'scripts/check-requirements.sh'
- 'convert*.py'
- 'requirements.txt'
- 'requirements/*.txt'
- '**/requirements*.txt'

concurrency:
group: ${{ github.workflow }}-${{ github.head_ref && github.ref || github.run_id }}
Expand Down
3 changes: 3 additions & 0 deletions .gitignore
Original file line number Diff line number Diff line change
Expand Up @@ -128,3 +128,6 @@ poetry.toml

# Scripts
!/scripts/install-oneapi.bat

# Test models for lora adapters
/lora-tests
5 changes: 4 additions & 1 deletion CMakePresets.json
Original file line number Diff line number Diff line change
Expand Up @@ -28,6 +28,7 @@
{ "name": "release", "hidden": true, "cacheVariables": { "CMAKE_BUILD_TYPE": "Release" } },
{ "name": "reldbg", "hidden": true, "cacheVariables": { "CMAKE_BUILD_TYPE": "RelWithDebInfo" } },
{ "name": "static", "hidden": true, "cacheVariables": { "GGML_STATIC": "ON" } },
{ "name": "sycl_f16", "hidden": true, "cacheVariables": { "GGML_SYCL_F16": "ON" } },

{
"name": "arm64-windows-msvc", "hidden": true,
Expand Down Expand Up @@ -60,6 +61,8 @@
{ "name": "x64-windows-msvc+static-release", "inherits": [ "base", "reldbg", "static" ] },

{ "name": "x64-windows-sycl-debug" , "inherits": [ "sycl-base", "debug" ] },
{ "name": "x64-windows-sycl-release", "inherits": [ "sycl-base", "release" ] }
{ "name": "x64-windows-sycl-debug-f16", "inherits": [ "sycl-base", "debug", "sycl_f16" ] },
{ "name": "x64-windows-sycl-release", "inherits": [ "sycl-base", "release" ] },
{ "name": "x64-windows-sycl-release-f16", "inherits": [ "sycl-base", "release", "sycl_f16" ] }
]
}
4 changes: 4 additions & 0 deletions Makefile
Original file line number Diff line number Diff line change
Expand Up @@ -763,6 +763,10 @@ ifdef GGML_VULKAN_MEMORY_DEBUG
MK_CPPFLAGS += -DGGML_VULKAN_MEMORY_DEBUG
endif

ifdef GGML_VULKAN_PERF
MK_CPPFLAGS += -DGGML_VULKAN_PERF
endif

ifdef GGML_VULKAN_VALIDATE
MK_CPPFLAGS += -DGGML_VULKAN_VALIDATE
endif
Expand Down
5 changes: 5 additions & 0 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -105,6 +105,8 @@ Typically finetunes of the base models below are supported as well.
- [x] [Open Elm models](https://huggingface.co/collections/apple/openelm-instruct-models-6619ad295d7ae9f868b759ca)
- [x] [ChatGLM3-6b](https://huggingface.co/THUDM/chatglm3-6b) + [ChatGLM4-9b](https://huggingface.co/THUDM/glm-4-9b)
- [x] [SmolLM](https://huggingface.co/collections/HuggingFaceTB/smollm-6695016cad7167254ce15966)
- [x] [EXAONE-3.0-7.8B-Instruct](https://huggingface.co/LGAI-EXAONE/EXAONE-3.0-7.8B-Instruct)
- [x] [FalconMamba Models](https://huggingface.co/collections/tiiuae/falconmamba-7b-66b9a580324dd1598b0f6d4a)

(instructions for supporting more models: [HOWTO-add-model.md](./docs/development/HOWTO-add-model.md))

Expand Down Expand Up @@ -186,10 +188,12 @@ Unless otherwise noted these projects are open-source with permissive licensing:

- [akx/ggify](https://github.com/akx/ggify) – download PyTorch models from HuggingFace Hub and convert them to GGML
- [crashr/gppm](https://github.com/crashr/gppm) – launch llama.cpp instances utilizing NVIDIA Tesla P40 or P100 GPUs with reduced idle power consumption
- [gpustack/gguf-parser](https://github.com/gpustack/gguf-parser-go/tree/main/cmd/gguf-parser) - review/check the GGUF file and estimate the memory usage

**Infrastructure:**

- [Paddler](https://github.com/distantmagic/paddler) - Stateful load balancer custom-tailored for llama.cpp
- [GPUStack](https://github.com/gpustack/gpustack) - Manage GPU clusters for running LLMs

**Games:**
- [Lucy's Labyrinth](https://github.com/MorganRO8/Lucys_Labyrinth) - A simple maze game where agents controlled by an AI model will try to trick you.
Expand Down Expand Up @@ -422,6 +426,7 @@ Please refer to [Build llama.cpp locally](./docs/build.md)
| [CUDA](./docs/build.md#cuda) | Nvidia GPU |
| [hipBLAS](./docs/build.md#hipblas) | AMD GPU |
| [Vulkan](./docs/build.md#vulkan) | GPU |
| [CANN](./docs/build.md#cann) | Ascend NPU |

## Tools

Expand Down
29 changes: 17 additions & 12 deletions ci/run.sh
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,9 @@
# # with SYCL support
# GG_BUILD_SYCL=1 bash ./ci/run.sh ./tmp/results ./tmp/mnt
#
# # with VULKAN support
# GG_BUILD_VULKAN=1 bash ./ci/run.sh ./tmp/results ./tmp/mnt
#

if [ -z "$2" ]; then
echo "usage: $0 <output-dir> <mnt-dir>"
Expand Down Expand Up @@ -40,7 +43,7 @@ if [ ! -z ${GG_BUILD_METAL} ]; then
fi

if [ ! -z ${GG_BUILD_CUDA} ]; then
CMAKE_EXTRA="${CMAKE_EXTRA} -DGGML_CUDA=1"
CMAKE_EXTRA="${CMAKE_EXTRA} -DGGML_CUDA=ON -DCMAKE_CUDA_ARCHITECTURES=native"
fi

if [ ! -z ${GG_BUILD_SYCL} ]; then
Expand All @@ -52,6 +55,10 @@ if [ ! -z ${GG_BUILD_SYCL} ]; then

CMAKE_EXTRA="${CMAKE_EXTRA} -DGGML_SYCL=1 DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DGGML_SYCL_F16=ON"
fi

if [ ! -z ${GG_BUILD_VULKAN} ]; then
CMAKE_EXTRA="${CMAKE_EXTRA} -DGGML_VULKAN=1"
fi
## helpers

# download a file if it does not exist or if it is outdated
Expand Down Expand Up @@ -107,7 +114,7 @@ function gg_run_ctest_debug {
gg_check_build_requirements

(time cmake -DCMAKE_BUILD_TYPE=Debug ${CMAKE_EXTRA} .. ) 2>&1 | tee -a $OUT/${ci}-cmake.log
(time make -j ) 2>&1 | tee -a $OUT/${ci}-make.log
(time make -j$(nproc) ) 2>&1 | tee -a $OUT/${ci}-make.log

(time ctest --output-on-failure -L main -E test-opt ) 2>&1 | tee -a $OUT/${ci}-ctest.log

Expand Down Expand Up @@ -138,7 +145,7 @@ function gg_run_ctest_release {
gg_check_build_requirements

(time cmake -DCMAKE_BUILD_TYPE=Release ${CMAKE_EXTRA} .. ) 2>&1 | tee -a $OUT/${ci}-cmake.log
(time make -j ) 2>&1 | tee -a $OUT/${ci}-make.log
(time make -j$(nproc) ) 2>&1 | tee -a $OUT/${ci}-make.log

if [ -z ${GG_BUILD_LOW_PERF} ]; then
(time ctest --output-on-failure -L main ) 2>&1 | tee -a $OUT/${ci}-ctest.log
Expand Down Expand Up @@ -266,7 +273,6 @@ function gg_sum_ctest_with_model_release {
}

# open_llama_7b_v2
# requires: GG_BUILD_CUDA

function gg_run_open_llama_7b_v2 {
cd ${SRC}
Expand All @@ -290,8 +296,8 @@ function gg_run_open_llama_7b_v2 {

set -e

(time cmake -DCMAKE_BUILD_TYPE=Release ${CMAKE_EXTRA} -DGGML_CUDA=1 .. ) 2>&1 | tee -a $OUT/${ci}-cmake.log
(time make -j ) 2>&1 | tee -a $OUT/${ci}-make.log
(time cmake -DCMAKE_BUILD_TYPE=Release ${CMAKE_EXTRA} .. ) 2>&1 | tee -a $OUT/${ci}-cmake.log
(time make -j$(nproc) ) 2>&1 | tee -a $OUT/${ci}-make.log

python3 ../examples/convert_legacy_llama.py ${path_models} --outfile ${path_models}/ggml-model-f16.gguf

Expand Down Expand Up @@ -425,7 +431,7 @@ function gg_run_pythia_1_4b {
set -e

(time cmake -DCMAKE_BUILD_TYPE=Release ${CMAKE_EXTRA} .. ) 2>&1 | tee -a $OUT/${ci}-cmake.log
(time make -j ) 2>&1 | tee -a $OUT/${ci}-make.log
(time make -j$(nproc) ) 2>&1 | tee -a $OUT/${ci}-make.log

python3 ../convert_hf_to_gguf.py ${path_models} --outfile ${path_models}/ggml-model-f16.gguf

Expand Down Expand Up @@ -535,7 +541,6 @@ function gg_sum_pythia_1_4b {
}

# pythia_2_8b
# requires: GG_BUILD_CUDA

function gg_run_pythia_2_8b {
cd ${SRC}
Expand All @@ -556,8 +561,8 @@ function gg_run_pythia_2_8b {

set -e

(time cmake -DCMAKE_BUILD_TYPE=Release ${CMAKE_EXTRA} -DGGML_CUDA=1 .. ) 2>&1 | tee -a $OUT/${ci}-cmake.log
(time make -j ) 2>&1 | tee -a $OUT/${ci}-make.log
(time cmake -DCMAKE_BUILD_TYPE=Release ${CMAKE_EXTRA} .. ) 2>&1 | tee -a $OUT/${ci}-cmake.log
(time make -j$(nproc) ) 2>&1 | tee -a $OUT/${ci}-make.log

python3 ../convert_hf_to_gguf.py ${path_models} --outfile ${path_models}/ggml-model-f16.gguf

Expand Down Expand Up @@ -692,7 +697,7 @@ function gg_run_embd_bge_small {
set -e

(time cmake -DCMAKE_BUILD_TYPE=Release ${CMAKE_EXTRA} .. ) 2>&1 | tee -a $OUT/${ci}-cmake.log
(time make -j ) 2>&1 | tee -a $OUT/${ci}-make.log
(time make -j$(nproc) ) 2>&1 | tee -a $OUT/${ci}-make.log

python3 ../convert_hf_to_gguf.py ${path_models} --outfile ${path_models}/ggml-model-f16.gguf

Expand Down Expand Up @@ -761,7 +766,7 @@ if [ -z ${GG_BUILD_LOW_PERF} ]; then
fi

if [ -z ${GG_BUILD_VRAM_GB} ] || [ ${GG_BUILD_VRAM_GB} -ge 8 ]; then
if [ -z ${GG_BUILD_CUDA} ]; then
if [ -z ${GG_BUILD_CUDA} ] && [ -z ${GG_BUILD_VULKAN} ]; then
test $ret -eq 0 && gg_run pythia_1_4b
else
test $ret -eq 0 && gg_run pythia_2_8b
Expand Down
Loading

0 comments on commit 3437c58

Please sign in to comment.