diff --git a/.devops/cann.Dockerfile b/.devops/cann.Dockerfile
index cd8f87b2ea..db221b0b81 100644
--- a/.devops/cann.Dockerfile
+++ b/.devops/cann.Dockerfile
@@ -4,7 +4,7 @@
# Define the CANN base image for easier version updates later
ARG CHIP_TYPE=910b
-ARG CANN_BASE_IMAGE=quay.io/ascend/cann:8.3.rc1.alpha001-${CHIP_TYPE}-openeuler22.03-py3.11
+ARG CANN_BASE_IMAGE=quay.io/ascend/cann:8.3.rc2-${CHIP_TYPE}-openeuler24.03-py3.11
# ==============================================================================
# BUILD STAGE
@@ -107,11 +107,11 @@ ENTRYPOINT ["/app/tools.sh"]
# ENTRYPOINT ["/app/llama-server"]
### Target: light
-# Lightweight image containing only llama-cli
+# Lightweight image containing only llama-cli and llama-completion
# ==============================================================================
FROM base AS light
-COPY --from=build /app/full/llama-cli /app
+COPY --from=build /app/full/llama-cli /app/full/llama-completion /app
ENTRYPOINT [ "/app/llama-cli" ]
diff --git a/.devops/cpu.Dockerfile b/.devops/cpu.Dockerfile
index 6e16ecda44..b9e84ab986 100644
--- a/.devops/cpu.Dockerfile
+++ b/.devops/cpu.Dockerfile
@@ -68,7 +68,7 @@ ENTRYPOINT ["/app/tools.sh"]
### Light, CLI only
FROM base AS light
-COPY --from=build /app/full/llama-cli /app
+COPY --from=build /app/full/llama-cli /app/full/llama-completion /app
WORKDIR /app
diff --git a/.devops/cuda-new.Dockerfile b/.devops/cuda-new.Dockerfile
new file mode 100644
index 0000000000..62443e17f2
--- /dev/null
+++ b/.devops/cuda-new.Dockerfile
@@ -0,0 +1,95 @@
+ARG UBUNTU_VERSION=24.04
+# This needs to generally match the container host's environment.
+ARG CUDA_VERSION=13.1.0
+# Target the CUDA build image
+ARG BASE_CUDA_DEV_CONTAINER=nvidia/cuda:${CUDA_VERSION}-devel-ubuntu${UBUNTU_VERSION}
+
+ARG BASE_CUDA_RUN_CONTAINER=nvidia/cuda:${CUDA_VERSION}-runtime-ubuntu${UBUNTU_VERSION}
+
+FROM ${BASE_CUDA_DEV_CONTAINER} AS build
+
+# CUDA architecture to build for (defaults to all supported archs)
+ARG CUDA_DOCKER_ARCH=default
+
+RUN apt-get update && \
+ apt-get install -y build-essential cmake python3 python3-pip git libcurl4-openssl-dev libgomp1
+
+WORKDIR /app
+
+COPY . .
+
+RUN if [ "${CUDA_DOCKER_ARCH}" != "default" ]; then \
+ export CMAKE_ARGS="-DCMAKE_CUDA_ARCHITECTURES=${CUDA_DOCKER_ARCH}"; \
+ fi && \
+ cmake -B build -DGGML_NATIVE=OFF -DGGML_CUDA=ON -DGGML_BACKEND_DL=ON -DGGML_CPU_ALL_VARIANTS=ON -DLLAMA_BUILD_TESTS=OFF ${CMAKE_ARGS} -DCMAKE_EXE_LINKER_FLAGS=-Wl,--allow-shlib-undefined . && \
+ cmake --build build --config Release -j$(nproc)
+
+RUN mkdir -p /app/lib && \
+ find build -name "*.so*" -exec cp -P {} /app/lib \;
+
+RUN mkdir -p /app/full \
+ && cp build/bin/* /app/full \
+ && cp *.py /app/full \
+ && cp -r gguf-py /app/full \
+ && cp -r requirements /app/full \
+ && cp requirements.txt /app/full \
+ && cp .devops/tools.sh /app/full/tools.sh
+
+## Base image
+FROM ${BASE_CUDA_RUN_CONTAINER} AS base
+
+RUN apt-get update \
+ && apt-get install -y libgomp1 curl\
+ && apt autoremove -y \
+ && apt clean -y \
+ && rm -rf /tmp/* /var/tmp/* \
+ && find /var/cache/apt/archives /var/lib/apt/lists -not -name lock -type f -delete \
+ && find /var/cache -type f -delete
+
+COPY --from=build /app/lib/ /app
+
+### Full
+FROM base AS full
+
+COPY --from=build /app/full /app
+
+WORKDIR /app
+
+RUN apt-get update \
+ && apt-get install -y \
+ git \
+ python3 \
+ python3-pip \
+ python3-wheel \
+ && pip install --break-system-packages --upgrade setuptools \
+ && pip install --break-system-packages -r requirements.txt \
+ && apt autoremove -y \
+ && apt clean -y \
+ && rm -rf /tmp/* /var/tmp/* \
+ && find /var/cache/apt/archives /var/lib/apt/lists -not -name lock -type f -delete \
+ && find /var/cache -type f -delete
+
+
+ENTRYPOINT ["/app/tools.sh"]
+
+### Light, CLI only
+FROM base AS light
+
+COPY --from=build /app/full/llama-cli /app/full/llama-completion /app
+
+WORKDIR /app
+
+ENTRYPOINT [ "/app/llama-cli" ]
+
+### Server, Server only
+FROM base AS server
+
+ENV LLAMA_ARG_HOST=0.0.0.0
+
+COPY --from=build /app/full/llama-server /app
+
+WORKDIR /app
+
+HEALTHCHECK CMD [ "curl", "-f", "http://localhost:8080/health" ]
+
+ENTRYPOINT [ "/app/llama-server" ]
diff --git a/.devops/cuda.Dockerfile b/.devops/cuda.Dockerfile
index 54f793d0a3..fed5863157 100644
--- a/.devops/cuda.Dockerfile
+++ b/.devops/cuda.Dockerfile
@@ -74,7 +74,7 @@ ENTRYPOINT ["/app/tools.sh"]
### Light, CLI only
FROM base AS light
-COPY --from=build /app/full/llama-cli /app
+COPY --from=build /app/full/llama-cli /app/full/llama-completion /app
WORKDIR /app
diff --git a/.devops/intel.Dockerfile b/.devops/intel.Dockerfile
index d1a8fbed4c..adebf08229 100644
--- a/.devops/intel.Dockerfile
+++ b/.devops/intel.Dockerfile
@@ -73,7 +73,7 @@ ENTRYPOINT ["/app/tools.sh"]
FROM base AS light
COPY --from=build /app/lib/ /app
-COPY --from=build /app/full/llama-cli /app
+COPY --from=build /app/full/llama-cli /app/full/llama-completion /app
WORKDIR /app
diff --git a/.devops/llama-cli-cann.Dockerfile b/.devops/llama-cli-cann.Dockerfile
index ef43d78cd2..6581187f32 100644
--- a/.devops/llama-cli-cann.Dockerfile
+++ b/.devops/llama-cli-cann.Dockerfile
@@ -23,11 +23,12 @@ ENV LD_LIBRARY_PATH=${ASCEND_TOOLKIT_HOME}/runtime/lib64/stub:$LD_LIBRARY_PATH
RUN echo "Building with static libs" && \
source /usr/local/Ascend/ascend-toolkit/set_env.sh --force && \
cmake -B build -DGGML_NATIVE=OFF -DGGML_CANN=ON -DBUILD_SHARED_LIBS=OFF -DLLAMA_BUILD_TESTS=OFF && \
- cmake --build build --config Release --target llama-cli
+ cmake --build build --config Release --target llama-cli && \
+ cmake --build build --config Release --target llama-completion
# TODO: use image with NNRT
FROM ascendai/cann:$ASCEND_VERSION AS runtime
-COPY --from=build /app/build/bin/llama-cli /llama-cli
+COPY --from=build /app/build/bin/llama-cli /app/build/bin/llama-completion /
ENV LC_ALL=C.utf8
diff --git a/.devops/llama-cpp-cuda.srpm.spec b/.devops/llama-cpp-cuda.srpm.spec
index 3bbf4a4def..4d42a906b1 100644
--- a/.devops/llama-cpp-cuda.srpm.spec
+++ b/.devops/llama-cpp-cuda.srpm.spec
@@ -37,6 +37,7 @@ make -j GGML_CUDA=1
%install
mkdir -p %{buildroot}%{_bindir}/
cp -p llama-cli %{buildroot}%{_bindir}/llama-cuda-cli
+cp -p llama-completion %{buildroot}%{_bindir}/llama-cuda-completion
cp -p llama-server %{buildroot}%{_bindir}/llama-cuda-server
cp -p llama-simple %{buildroot}%{_bindir}/llama-cuda-simple
@@ -68,6 +69,7 @@ rm -rf %{_builddir}/*
%files
%{_bindir}/llama-cuda-cli
+%{_bindir}/llama-cuda-completion
%{_bindir}/llama-cuda-server
%{_bindir}/llama-cuda-simple
/usr/lib/systemd/system/llamacuda.service
diff --git a/.devops/llama-cpp.srpm.spec b/.devops/llama-cpp.srpm.spec
index 45902dcf89..0a4f43058d 100644
--- a/.devops/llama-cpp.srpm.spec
+++ b/.devops/llama-cpp.srpm.spec
@@ -39,6 +39,7 @@ make -j
%install
mkdir -p %{buildroot}%{_bindir}/
cp -p llama-cli %{buildroot}%{_bindir}/llama-cli
+cp -p llama-completion %{buildroot}%{_bindir}/llama-completion
cp -p llama-server %{buildroot}%{_bindir}/llama-server
cp -p llama-simple %{buildroot}%{_bindir}/llama-simple
@@ -70,6 +71,7 @@ rm -rf %{_builddir}/*
%files
%{_bindir}/llama-cli
+%{_bindir}/llama-completion
%{_bindir}/llama-server
%{_bindir}/llama-simple
/usr/lib/systemd/system/llama.service
diff --git a/.devops/musa.Dockerfile b/.devops/musa.Dockerfile
index faa3500e61..34d6ad9f40 100644
--- a/.devops/musa.Dockerfile
+++ b/.devops/musa.Dockerfile
@@ -81,7 +81,7 @@ ENTRYPOINT ["/app/tools.sh"]
### Light, CLI only
FROM base AS light
-COPY --from=build /app/full/llama-cli /app
+COPY --from=build /app/full/llama-cli /app/full/llama-completion /app
WORKDIR /app
diff --git a/.devops/rocm.Dockerfile b/.devops/rocm.Dockerfile
index d6bf28b105..53c3ed8d88 100644
--- a/.devops/rocm.Dockerfile
+++ b/.devops/rocm.Dockerfile
@@ -94,7 +94,7 @@ ENTRYPOINT ["/app/tools.sh"]
### Light, CLI only
FROM base AS light
-COPY --from=build /app/full/llama-cli /app
+COPY --from=build /app/full/llama-cli /app/full/llama-completion /app
WORKDIR /app
diff --git a/.devops/s390x.Dockerfile b/.devops/s390x.Dockerfile
index b7c9457680..1e66f061d5 100644
--- a/.devops/s390x.Dockerfile
+++ b/.devops/s390x.Dockerfile
@@ -105,7 +105,7 @@ WORKDIR /llama.cpp/bin
# Copy llama.cpp binaries and libraries
COPY --from=collector /llama.cpp/bin/*.so /llama.cpp/bin
-COPY --from=collector /llama.cpp/bin/llama-cli /llama.cpp/bin
+COPY --from=collector /llama.cpp/bin/llama-cli /llama.cpp/bin/llama-completion /llama.cpp/bin
ENTRYPOINT [ "/llama.cpp/bin/llama-cli" ]
diff --git a/.devops/tools.sh b/.devops/tools.sh
index 8a3a693400..cc5ee17dfd 100755
--- a/.devops/tools.sh
+++ b/.devops/tools.sh
@@ -13,6 +13,8 @@ elif [[ "$arg1" == '--quantize' || "$arg1" == '-q' ]]; then
exec ./llama-quantize "$@"
elif [[ "$arg1" == '--run' || "$arg1" == '-r' ]]; then
exec ./llama-cli "$@"
+elif [[ "$arg1" == '--run-legacy' || "$arg1" == '-l' ]]; then
+ exec ./llama-completion "$@"
elif [[ "$arg1" == '--bench' || "$arg1" == '-b' ]]; then
exec ./llama-bench "$@"
elif [[ "$arg1" == '--perplexity' || "$arg1" == '-p' ]]; then
@@ -32,8 +34,10 @@ elif [[ "$arg1" == '--server' || "$arg1" == '-s' ]]; then
else
echo "Unknown command: $arg1"
echo "Available commands: "
- echo " --run (-r): Run a model previously converted into ggml"
- echo " ex: -m /models/7B/ggml-model-q4_0.bin -p \"Building a website can be done in 10 simple steps:\" -n 512"
+ echo " --run (-r): Run a model (chat) previously converted into ggml"
+ echo " ex: -m /models/7B/ggml-model-q4_0.bin"
+ echo " --run-legacy (-l): Run a model (legacy completion) previously converted into ggml"
+ echo " ex: -m /models/7B/ggml-model-q4_0.bin -no-cnv -p \"Building a website can be done in 10 simple steps:\" -n 512"
echo " --bench (-b): Benchmark the performance of the inference for various parameters."
echo " ex: -m model.gguf"
echo " --perplexity (-p): Measure the perplexity of a model over a given text."
diff --git a/.devops/vulkan.Dockerfile b/.devops/vulkan.Dockerfile
index fd7195c5be..b37b4f277d 100644
--- a/.devops/vulkan.Dockerfile
+++ b/.devops/vulkan.Dockerfile
@@ -68,7 +68,7 @@ ENTRYPOINT ["/app/tools.sh"]
### Light, CLI only
FROM base AS light
-COPY --from=build /app/full/llama-cli /app
+COPY --from=build /app/full/llama-cli /app/full/llama-completion /app
WORKDIR /app
diff --git a/.gemini/settings.json b/.gemini/settings.json
new file mode 100644
index 0000000000..68337d390f
--- /dev/null
+++ b/.gemini/settings.json
@@ -0,0 +1 @@
+{ "contextFileName": "AGENTS.md" }
diff --git a/.github/ISSUE_TEMPLATE/010-bug-compilation.yml b/.github/ISSUE_TEMPLATE/010-bug-compilation.yml
index feb0d51205..c106f47a25 100644
--- a/.github/ISSUE_TEMPLATE/010-bug-compilation.yml
+++ b/.github/ISSUE_TEMPLATE/010-bug-compilation.yml
@@ -8,7 +8,8 @@ body:
value: >
Thanks for taking the time to fill out this bug report!
This issue template is intended for bug reports where the compilation of llama.cpp fails.
- Before opening an issue, please confirm that the compilation still fails with `-DGGML_CCACHE=OFF`.
+ Before opening an issue, please confirm that the compilation still fails
+ after recreating the CMake build directory and with `-DGGML_CCACHE=OFF`.
If the compilation succeeds with ccache disabled you should be able to permanently fix the issue
by clearing `~/.cache/ccache` (on Linux).
- type: textarea
diff --git a/.github/ISSUE_TEMPLATE/011-bug-results.yml b/.github/ISSUE_TEMPLATE/011-bug-results.yml
index c42a14ff83..31202dfa83 100644
--- a/.github/ISSUE_TEMPLATE/011-bug-results.yml
+++ b/.github/ISSUE_TEMPLATE/011-bug-results.yml
@@ -11,7 +11,7 @@ body:
(i.e. the generated text) are incorrect or llama.cpp crashes during model evaluation.
If you encountered the issue while using an external UI (e.g. ollama),
please reproduce your issue using one of the examples/binaries in this repository.
- The `llama-cli` binary can be used for simple and reproducible model inference.
+ The `llama-completion` binary can be used for simple and reproducible model inference.
- type: textarea
id: version
attributes:
@@ -74,9 +74,12 @@ body:
Please give us a summary of the problem and tell us how to reproduce it.
If you can narrow down the bug to specific hardware, compile flags, or command line arguments,
that information would be very much appreciated by us.
+
+ If possible, please try to reproduce the issue using `llama-completion` with `-fit off`.
+ If you can only reproduce the issue with `-fit on`, please provide logs both with and without `--verbose`.
placeholder: >
- e.g. when I run llama-cli with -ngl 99 I get garbled outputs.
- When I use -ngl 0 it works correctly.
+ e.g. when I run llama-completion with `-fa on` I get garbled outputs for very long prompts.
+ With short prompts or `-fa off` it works correctly.
Here are the exact commands that I used: ...
validations:
required: true
@@ -95,7 +98,18 @@ body:
label: Relevant log output
description: >
Please copy and paste any relevant log output, including the command that you entered and any generated text.
- This will be automatically formatted into code, so no need for backticks.
- render: shell
+ For very long logs (thousands of lines), preferably upload them as files instead.
+ On Linux you can redirect console output into a file by appending ` > llama.log 2>&1` to your command.
+ value: |
+
+ Logs
+
+
+ ```console
+
+ ```
+
+
+
validations:
required: true
diff --git a/.github/ISSUE_TEMPLATE/019-bug-misc.yml b/.github/ISSUE_TEMPLATE/019-bug-misc.yml
index 1904e31fdc..8e867e7f60 100644
--- a/.github/ISSUE_TEMPLATE/019-bug-misc.yml
+++ b/.github/ISSUE_TEMPLATE/019-bug-misc.yml
@@ -85,7 +85,19 @@ body:
label: Relevant log output
description: >
If applicable, please copy and paste any relevant log output, including any generated text.
- This will be automatically formatted into code, so no need for backticks.
- render: shell
+ If you are encountering problems specifically with the `llama_params_fit` module, always upload `--verbose` logs as well.
+ For very long logs (thousands of lines), please upload them as files instead.
+ On Linux you can redirect console output into a file by appending ` > llama.log 2>&1` to your command.
+ value: |
+
+ Logs
+
+
+ ```console
+
+ ```
+
+
+
validations:
required: false
diff --git a/.github/actions/windows-setup-cuda/action.yml b/.github/actions/windows-setup-cuda/action.yml
index 5575caeca3..6ad61582a5 100644
--- a/.github/actions/windows-setup-cuda/action.yml
+++ b/.github/actions/windows-setup-cuda/action.yml
@@ -65,3 +65,34 @@ runs:
echo "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4\libnvvp" | Out-File -FilePath $env:GITHUB_PATH -Encoding utf8 -Append
echo "CUDA_PATH=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4" | Out-File -FilePath $env:GITHUB_ENV -Append -Encoding utf8
echo "CUDA_PATH_V12_4=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4" | Out-File -FilePath $env:GITHUB_ENV -Append -Encoding utf8
+
+ - name: Install Cuda Toolkit 13.1
+ if: ${{ inputs.cuda_version == '13.1' }}
+ shell: pwsh
+ run: |
+ mkdir -p "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1"
+ choco install unzip -y
+ curl -O "https://developer.download.nvidia.com/compute/cuda/redist/cuda_crt/windows-x86_64/cuda_crt-windows-x86_64-13.1.80-archive.zip"
+ curl -O "https://developer.download.nvidia.com/compute/cuda/redist/cuda_cudart/windows-x86_64/cuda_cudart-windows-x86_64-13.1.80-archive.zip"
+ curl -O "https://developer.download.nvidia.com/compute/cuda/redist/cuda_nvcc/windows-x86_64/cuda_nvcc-windows-x86_64-13.1.80-archive.zip"
+ curl -O "https://developer.download.nvidia.com/compute/cuda/redist/cuda_nvrtc/windows-x86_64/cuda_nvrtc-windows-x86_64-13.1.80-archive.zip"
+ curl -O "https://developer.download.nvidia.com/compute/cuda/redist/libcublas/windows-x86_64/libcublas-windows-x86_64-13.2.0.9-archive.zip"
+ curl -O "https://developer.download.nvidia.com/compute/cuda/redist/libnvvm/windows-x86_64/libnvvm-windows-x86_64-13.1.80-archive.zip"
+ curl -O "https://developer.download.nvidia.com/compute/cuda/redist/cuda_nvtx/windows-x86_64/cuda_nvtx-windows-x86_64-13.1.68-archive.zip"
+ curl -O "https://developer.download.nvidia.com/compute/cuda/redist/cuda_profiler_api/windows-x86_64/cuda_profiler_api-windows-x86_64-13.1.80-archive.zip"
+ curl -O "https://developer.download.nvidia.com/compute/cuda/redist/visual_studio_integration/windows-x86_64/visual_studio_integration-windows-x86_64-13.1.68-archive.zip"
+ curl -O "https://developer.download.nvidia.com/compute/cuda/redist/cuda_cccl/windows-x86_64/cuda_cccl-windows-x86_64-13.1.78-archive.zip"
+ unzip '*.zip' -d "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1"
+ xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1\cuda_crt-windows-x86_64-13.1.80-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1" /E /I /H /Y
+ xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1\cuda_cudart-windows-x86_64-13.1.80-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1" /E /I /H /Y
+ xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1\cuda_nvcc-windows-x86_64-13.1.80-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1" /E /I /H /Y
+ xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1\cuda_nvrtc-windows-x86_64-13.1.80-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1" /E /I /H /Y
+ xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1\libcublas-windows-x86_64-13.2.0.9-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1" /E /I /H /Y
+ xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1\libnvvm-windows-x86_64-13.1.80-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1" /E /I /H /Y
+ xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1\cuda_nvtx-windows-x86_64-13.1.68-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1" /E /I /H /Y
+ xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1\cuda_profiler_api-windows-x86_64-13.1.80-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1" /E /I /H /Y
+ xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1\visual_studio_integration-windows-x86_64-13.1.68-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1" /E /I /H /Y
+ xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1\cuda_cccl-windows-x86_64-13.1.78-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1" /E /I /H /Y
+ echo "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1\bin" | Out-File -FilePath $env:GITHUB_PATH -Encoding utf8 -Append
+ echo "CUDA_PATH=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1" | Out-File -FilePath $env:GITHUB_ENV -Append -Encoding utf8
+ echo "CUDA_PATH_V13_1=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.1" | Out-File -FilePath $env:GITHUB_ENV -Append -Encoding utf8
diff --git a/.github/copilot-instructions.md b/.github/copilot-instructions.md
deleted file mode 100644
index ad13c6ea8d..0000000000
--- a/.github/copilot-instructions.md
+++ /dev/null
@@ -1,262 +0,0 @@
-# Copilot Instructions for llama.cpp
-
-## Repository Overview
-
-llama.cpp is a large-scale C/C++ project for efficient LLM (Large Language Model) inference with minimal setup and dependencies. The project enables running language models on diverse hardware with state-of-the-art performance.
-
-**Key Facts:**
-- **Primary language**: C/C++ with Python utility scripts
-- **Size**: ~200k+ lines of code across 1000+ files
-- **Architecture**: Modular design with main library (`libllama`) and 40+ executable tools/examples
-- **Core dependency**: ggml tensor library (vendored in `ggml/` directory)
-- **Backends supported**: CPU (AVX/NEON/RVV optimized), CUDA, Metal, Vulkan, SYCL, ROCm, MUSA
-- **License**: MIT
-
-## Build Instructions
-
-### Prerequisites
-- CMake 3.14+ (primary build system)
-- C++17 compatible compiler (GCC 13.3+, Clang, MSVC)
-- Optional: ccache for faster compilation
-
-### Basic Build (CPU-only)
-**ALWAYS run these commands in sequence:**
-```bash
-cmake -B build
-cmake --build build --config Release -j $(nproc)
-```
-
-**Build time**: ~10 minutes on 4-core system with ccache enabled, ~25 minutes without ccache.
-
-**Important Notes:**
-- The Makefile is deprecated - always use CMake
-- ccache is automatically detected and used if available
-- Built binaries are placed in `build/bin/`
-- Parallel builds (`-j`) significantly reduce build time
-
-### Backend-Specific Builds
-For CUDA support:
-```bash
-cmake -B build -DGGML_CUDA=ON
-cmake --build build --config Release -j $(nproc)
-```
-
-For Metal (macOS):
-```bash
-cmake -B build -DGGML_METAL=ON
-cmake --build build --config Release -j $(nproc)
-```
-
-**Important Note**: While all backends can be built as long as the correct requirements for that backend are installed, you will not be able to run them without the correct hardware. The only backend that can be run for testing and validation is the CPU backend.
-
-### Debug Builds
-Single-config generators:
-```bash
-cmake -B build -DCMAKE_BUILD_TYPE=Debug
-cmake --build build
-```
-
-Multi-config generators:
-```bash
-cmake -B build -G "Xcode"
-cmake --build build --config Debug
-```
-
-### Common Build Issues
-- **Issue**: Network tests fail in isolated environments
- **Solution**: Expected behavior - core functionality tests will still pass
-
-## Testing
-
-### Running Tests
-```bash
-ctest --test-dir build --output-on-failure -j $(nproc)
-```
-
-**Test suite**: 38 tests covering tokenizers, grammar parsing, sampling, backends, and integration
-**Expected failures**: 2-3 tests may fail if network access is unavailable (they download models)
-**Test time**: ~30 seconds for passing tests
-
-### Server Unit Tests
-Run server-specific unit tests after building the server:
-```bash
-# Build the server first
-cmake --build build --target llama-server
-
-# Navigate to server tests and run
-cd tools/server/tests
-source ../../../.venv/bin/activate
-./tests.sh
-```
-**Server test dependencies**: The `.venv` environment includes the required dependencies for server unit tests (pytest, aiohttp, etc.). Tests can be run individually or with various options as documented in `tools/server/tests/README.md`.
-
-### Test Categories
-- Tokenizer tests: Various model tokenizers (BERT, GPT-2, LLaMA, etc.)
-- Grammar tests: GBNF parsing and validation
-- Backend tests: Core ggml operations across different backends
-- Integration tests: End-to-end workflows
-
-### Manual Testing Commands
-```bash
-# Test basic inference
-./build/bin/llama-cli --version
-
-# Test model loading (requires model file)
-./build/bin/llama-cli -m path/to/model.gguf -p "Hello" -n 10
-```
-
-## Code Quality and Linting
-
-### C++ Code Formatting
-**ALWAYS format C++ code before committing:**
-```bash
-git clang-format
-```
-
-Configuration is in `.clang-format` with these key rules:
-- 4-space indentation
-- 120 column limit
-- Braces on same line for functions
-- Pointer alignment: `void * ptr` (middle)
-- Reference alignment: `int & ref` (middle)
-
-### Python Code
-**ALWAYS activate the Python environment in `.venv` and use tools from that environment:**
-```bash
-# Activate virtual environment
-source .venv/bin/activate
-```
-
-Configuration files:
-- `.flake8`: flake8 settings (max-line-length=125, excludes examples/tools)
-- `pyrightconfig.json`: pyright type checking configuration
-
-### Pre-commit Hooks
-Run before committing:
-```bash
-pre-commit run --all-files
-```
-
-## Continuous Integration
-
-### GitHub Actions Workflows
-Key workflows that run on every PR:
-- `.github/workflows/build.yml`: Multi-platform builds
-- `.github/workflows/server.yml`: Server functionality tests
-- `.github/workflows/python-lint.yml`: Python code quality
-- `.github/workflows/python-type-check.yml`: Python type checking
-
-### Local CI Validation
-**Run full CI locally before submitting PRs:**
-```bash
-mkdir tmp
-
-# CPU-only build
-bash ./ci/run.sh ./tmp/results ./tmp/mnt
-```
-
-**CI Runtime**: 30-60 minutes depending on backend configuration
-
-### Triggering CI
-Add `ggml-ci` to commit message to trigger heavy CI workloads on the custom CI infrastructure.
-
-## Project Layout and Architecture
-
-### Core Directories
-- **`src/`**: Main llama library implementation (`llama.cpp`, `llama-*.cpp`)
-- **`include/`**: Public API headers, primarily `include/llama.h`
-- **`ggml/`**: Core tensor library (submodule with custom GGML framework)
-- **`examples/`**: 30+ example applications and tools
-- **`tools/`**: Additional development and utility tools (server benchmarks, tests)
-- **`tests/`**: Comprehensive test suite with CTest integration
-- **`docs/`**: Detailed documentation (build guides, API docs, etc.)
-- **`scripts/`**: Utility scripts for CI, data processing, and automation
-- **`common/`**: Shared utility code used across examples
-
-### Key Files
-- **`CMakeLists.txt`**: Primary build configuration
-- **`include/llama.h`**: Main C API header (~2000 lines)
-- **`src/llama.cpp`**: Core library implementation (~8000 lines)
-- **`CONTRIBUTING.md`**: Coding guidelines and PR requirements
-- **`.clang-format`**: C++ formatting rules
-- **`.pre-commit-config.yaml`**: Git hook configuration
-
-### Built Executables (in `build/bin/`)
-Primary tools:
-- **`llama-cli`**: Main inference tool
-- **`llama-server`**: OpenAI-compatible HTTP server
-- **`llama-quantize`**: Model quantization utility
-- **`llama-perplexity`**: Model evaluation tool
-- **`llama-bench`**: Performance benchmarking
-- **`llama-convert-llama2c-to-ggml`**: Model conversion utilities
-
-### Configuration Files
-- **CMake**: `CMakeLists.txt`, `cmake/` directory
-- **Linting**: `.clang-format`, `.clang-tidy`, `.flake8`
-- **CI**: `.github/workflows/`, `ci/run.sh`
-- **Git**: `.gitignore` (includes build artifacts, models, cache)
-
-### Dependencies
-- **System**: OpenMP, libcurl (for model downloading)
-- **Optional**: CUDA SDK, Metal framework, Vulkan SDK, Intel oneAPI
-- **Bundled**: httplib, json (header-only libraries in vendored form)
-
-## Common Validation Steps
-
-### After Making Changes
-1. **Format code**: `git clang-format`
-2. **Build**: `cmake --build build --config Release`
-3. **Test**: `ctest --test-dir build --output-on-failure`
-4. **Server tests** (if modifying server): `cd tools/server/tests && source ../../../.venv/bin/activate && ./tests.sh`
-5. **Manual validation**: Test relevant tools in `build/bin/`
-
-### Performance Validation
-```bash
-# Benchmark inference performance
-./build/bin/llama-bench -m model.gguf
-
-# Evaluate model perplexity
-./build/bin/llama-perplexity -m model.gguf -f dataset.txt
-```
-
-### Backend Validation
-```bash
-# Test backend operations
-./build/bin/test-backend-ops
-```
-
-## Environment Setup
-
-### Required Tools
-- CMake 3.14+ (install via system package manager)
-- Modern C++ compiler with C++17 support
-- Git (for submodule management)
-- Python 3.9+ with virtual environment (`.venv` is provided)
-
-### Optional but Recommended
-- ccache: `apt install ccache` or `brew install ccache`
-- clang-format 15+: Usually included with LLVM/Clang installation
-- pre-commit: `pip install pre-commit`
-
-### Backend-Specific Requirements
-- **CUDA**: NVIDIA CUDA Toolkit 11.2+
-- **Metal**: Xcode command line tools (macOS only)
-- **Vulkan**: Vulkan SDK
-- **SYCL**: Intel oneAPI toolkit
-
-## Important Guidelines
-
-### Code Changes
-- **Minimal dependencies**: Avoid adding new external dependencies
-- **Cross-platform compatibility**: Test on Linux, macOS, Windows when possible
-- **Performance focus**: This is a performance-critical inference library
-- **API stability**: Changes to `include/llama.h` require careful consideration
-
-### Git Workflow
-- Always create feature branches from `master`
-- **Never** commit build artifacts (`build/`, `.ccache/`, `*.o`, `*.gguf`)
-- Use descriptive commit messages following project conventions
-
-### Trust These Instructions
-Only search for additional information if these instructions are incomplete or found to be incorrect. This document contains validated build and test procedures that work reliably across different environments.
-
diff --git a/.github/workflows/build-linux-cross.yml b/.github/workflows/build-linux-cross.yml
index 36201281f0..c2c6ea12ae 100644
--- a/.github/workflows/build-linux-cross.yml
+++ b/.github/workflows/build-linux-cross.yml
@@ -291,6 +291,7 @@ jobs:
-DGGML_RVV=ON \
-DGGML_RV_ZFH=ON \
-DGGML_RV_ZICBOP=ON \
+ -DGGML_RV_ZIHINTPAUSE=ON \
-DRISCV64_SPACEMIT_IME_SPEC=RISCV64_SPACEMIT_IME1 \
-DCMAKE_TOOLCHAIN_FILE=${PWD}/cmake/riscv64-spacemit-linux-gnu-gcc.cmake
diff --git a/.github/workflows/build-riscv-native.yml b/.github/workflows/build-riscv-native.yml
deleted file mode 100644
index a3a0b0d663..0000000000
--- a/.github/workflows/build-riscv-native.yml
+++ /dev/null
@@ -1,120 +0,0 @@
-name: Build on RISCV Linux Machine by Cloud-V
-on:
- pull_request:
- workflow_dispatch:
- workflow_call:
-
-jobs:
- debian-13-riscv64-native: # Bianbu 2.2
- runs-on: [self-hosted, RISCV64]
-
- steps:
- - name: Install prerequisites
- run: |
- sudo apt-get update || true
- sudo apt-get install -y libatomic1
- - uses: actions/checkout@v4
- - name: Setup Riscv
- run: |
- sudo apt-get update || true
- sudo apt-get install -y --no-install-recommends \
- build-essential \
- gcc-14-riscv64-linux-gnu \
- g++-14-riscv64-linux-gnu \
- ccache \
- cmake
-
- - name: Setup ccache
- run: |
- mkdir -p $HOME/.ccache
- ccache -M 5G -d $HOME/.ccache
- export CCACHE_LOGFILE=/home/runneruser/ccache_debug/ccache.log
- export CCACHE_DEBUGDIR="/home/runneruser/ccache_debug"
- echo "$GITHUB_WORKSPACE"
- echo "CCACHE_LOGFILE=$CCACHE_LOGFILE" >> $GITHUB_ENV
- echo "CCACHE_DEBUGDIR=$CCACHE_DEBUGDIR" >> $GITHUB_ENV
- echo "CCACHE_BASEDIR=$GITHUB_WORKSPACE" >> $GITHUB_ENV
- echo "CCACHE_DIR=$HOME/.ccache" >> $GITHUB_ENV
-
- - name: Build
- run: |
- cmake -B build \
- -DLLAMA_CURL=OFF \
- -DCMAKE_BUILD_TYPE=Release \
- -DGGML_OPENMP=OFF \
- -DLLAMA_BUILD_EXAMPLES=ON \
- -DLLAMA_BUILD_TOOLS=ON \
- -DLLAMA_BUILD_TESTS=OFF \
- -DCMAKE_SYSTEM_NAME=Linux \
- -DCMAKE_SYSTEM_PROCESSOR=riscv64 \
- -DCMAKE_C_COMPILER=riscv64-linux-gnu-gcc-14 \
- -DCMAKE_CXX_COMPILER=riscv64-linux-gnu-g++-14 \
- -DCMAKE_C_COMPILER_LAUNCHER=ccache \
- -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
- -DCMAKE_POSITION_INDEPENDENT_CODE=ON \
- -DCMAKE_FIND_ROOT_PATH=/usr/lib/riscv64-linux-gnu \
- -DCMAKE_FIND_ROOT_PATH_MODE_PROGRAM=NEVER \
- -DCMAKE_FIND_ROOT_PATH_MODE_LIBRARY=ONLY \
- -DCMAKE_FIND_ROOT_PATH_MODE_INCLUDE=BOTH
-
- cmake --build build --config Release -j $(nproc)
-
- # debian-13-riscv64-spacemit-ime-native: # Bianbu 2.2
- # runs-on: [self-hosted, RISCV64]
-
- # steps:
- # - name: Install prerequisites
- # run: |
- # sudo apt-get update || true
- # sudo apt-get install -y libatomic1
- # - uses: actions/checkout@v4
- # - name: Setup Riscv
- # run: |
- # sudo apt-get update || true
- # sudo apt-get install -y --no-install-recommends \
- # build-essential \
- # gcc-14-riscv64-linux-gnu \
- # g++-14-riscv64-linux-gnu \
- # ccache \
- # cmake
- # sudo apt-get upgrade binutils -y
-
- # - name: Setup ccache
- # run: |
- # mkdir -p $HOME/.ccache
- # ccache -M 5G -d $HOME/.ccache
- # export CCACHE_LOGFILE=/home/runneruser/ccache_debug/ccache.log
- # export CCACHE_DEBUGDIR="/home/runneruser/ccache_debug"
- # echo "$GITHUB_WORKSPACE"
- # echo "CCACHE_LOGFILE=$CCACHE_LOGFILE" >> $GITHUB_ENV
- # echo "CCACHE_DEBUGDIR=$CCACHE_DEBUGDIR" >> $GITHUB_ENV
- # echo "CCACHE_BASEDIR=$GITHUB_WORKSPACE" >> $GITHUB_ENV
- # echo "CCACHE_DIR=$HOME/.ccache" >> $GITHUB_ENV
-
- # - name: Build
- # run: |
- # cmake -B build \
- # -DLLAMA_CURL=OFF \
- # -DCMAKE_BUILD_TYPE=Release \
- # -DGGML_OPENMP=OFF \
- # -DLLAMA_BUILD_EXAMPLES=ON \
- # -DLLAMA_BUILD_TOOLS=ON \
- # -DLLAMA_BUILD_TESTS=OFF \
- # -DCMAKE_SYSTEM_NAME=Linux \
- # -DCMAKE_SYSTEM_PROCESSOR=riscv64 \
- # -DCMAKE_C_COMPILER=riscv64-linux-gnu-gcc-14 \
- # -DCMAKE_CXX_COMPILER=riscv64-linux-gnu-g++-14 \
- # -DCMAKE_C_COMPILER_LAUNCHER=ccache \
- # -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
- # -DCMAKE_POSITION_INDEPENDENT_CODE=ON \
- # -DCMAKE_FIND_ROOT_PATH=/usr/lib/riscv64-linux-gnu \
- # -DCMAKE_FIND_ROOT_PATH_MODE_PROGRAM=NEVER \
- # -DCMAKE_FIND_ROOT_PATH_MODE_LIBRARY=ONLY \
- # -DCMAKE_FIND_ROOT_PATH_MODE_INCLUDE=BOTH \
- # -DGGML_RVV=ON \
- # -DGGML_RV_ZFH=ON \
- # -DGGML_RV_ZICBOP=ON \
- # -DGGML_CPU_RISCV64_SPACEMIT=ON \
- # -DRISCV64_SPACEMIT_IME_SPEC=RISCV64_SPACEMIT_IME1
-
- # cmake --build build --config Release -j $(nproc)
diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
index eee42759fc..1193779d0b 100644
--- a/.github/workflows/build.yml
+++ b/.github/workflows/build.yml
@@ -20,7 +20,8 @@ on:
'**/*.swift',
'**/*.m',
'**/*.metal',
- '**/*.comp'
+ '**/*.comp',
+ '**/*.glsl'
]
pull_request:
@@ -40,7 +41,8 @@ on:
'**/*.swift',
'**/*.m',
'**/*.metal',
- '**/*.comp'
+ '**/*.comp',
+ '**/*.glsl'
]
concurrency:
@@ -68,6 +70,7 @@ jobs:
with:
key: macOS-latest-cmake-arm64
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Build
id: cmake_build
@@ -104,6 +107,7 @@ jobs:
with:
key: macOS-latest-cmake-x64
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Build
id: cmake_build
@@ -140,6 +144,7 @@ jobs:
with:
key: macOS-latest-cmake-arm64-webgpu
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Dawn Dependency
id: dawn-depends
@@ -193,6 +198,7 @@ jobs:
with:
key: ubuntu-cpu-cmake-${{ matrix.build }}
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Build Dependencies
id: build_depends
@@ -243,7 +249,7 @@ jobs:
echo "Fetch llama2c model"
wget https://huggingface.co/karpathy/tinyllamas/resolve/main/stories260K/stories260K.bin
./bin/llama-convert-llama2c-to-ggml --copy-vocab-from-model ./tok512.bin --llama2c-model stories260K.bin --llama2c-output-model stories260K.gguf
- ./bin/llama-cli -m stories260K.gguf -p "One day, Lily met a Shoggoth" -n 500 -c 256
+ ./bin/llama-completion -m stories260K.gguf -p "One day, Lily met a Shoggoth" -n 500 -c 256
- name: Test llama2c (s390x)
id: llama2c_test_s390x
@@ -252,7 +258,7 @@ jobs:
cd build
echo "Fetch llama2c big-endian model"
wget https://huggingface.co/ggml-org/models/resolve/main/tinyllamas/stories260K-be.gguf
- ./bin/llama-cli -m stories260K-be.gguf -p "One day, Lily met a Shoggoth" -n 500 -c 256
+ ./bin/llama-completion -m stories260K-be.gguf -p "One day, Lily met a Shoggoth" -n 500 -c 256
ubuntu-latest-cmake-sanitizer:
runs-on: ubuntu-latest
@@ -274,6 +280,7 @@ jobs:
with:
key: ubuntu-latest-cmake-sanitizer-${{ matrix.sanitizer }}
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Dependencies
id: depends
@@ -394,6 +401,7 @@ jobs:
with:
key: ubuntu-24-cmake-vulkan-deb
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Dependencies
id: depends
@@ -429,6 +437,7 @@ jobs:
with:
key: ubuntu-24-cmake-vulkan
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Dependencies
id: depends
@@ -488,6 +497,7 @@ jobs:
with:
key: ubuntu-24-cmake-webgpu
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Dependencies
id: depends
@@ -547,6 +557,47 @@ jobs:
# This is using llvmpipe and runs slower than other backends
ctest -L main --verbose --timeout 3600
+ ubuntu-24-wasm-webgpu:
+ runs-on: ubuntu-24.04
+
+ steps:
+ - name: Clone
+ id: checkout
+ uses: actions/checkout@v4
+
+ - name: ccache
+ uses: ggml-org/ccache-action@v1.2.16
+ with:
+ key: ubuntu-latest-wasm-webgpu
+ evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
+
+ - name: Install Emscripten
+ run: |
+ git clone https://github.com/emscripten-core/emsdk.git
+ cd emsdk
+ ./emsdk install latest
+ ./emsdk activate latest
+
+ - name: Fetch emdawnwebgpu
+ run: |
+ DAWN_TAG="v20251027.212519"
+ EMDAWN_PKG="emdawnwebgpu_pkg-${DAWN_TAG}.zip"
+ echo "Downloading ${EMDAWN_PKG}"
+ curl -L -o emdawn.zip \
+ "https://github.com/google/dawn/releases/download/${DAWN_TAG}/${EMDAWN_PKG}"
+ unzip emdawn.zip
+
+ - name: Build WASM WebGPU
+ run: |
+ source emsdk/emsdk_env.sh
+ emcmake cmake -B build-wasm \
+ -DGGML_WEBGPU=ON \
+ -DLLAMA_CURL=OFF \
+ -DEMDAWNWEBGPU_DIR=emdawnwebgpu_pkg
+
+ cmake --build build-wasm --target test-backend-ops -j $(nproc)
+
ubuntu-22-cmake-hip:
runs-on: ubuntu-22.04
container: rocm/dev-ubuntu-22.04:6.1.2
@@ -567,6 +618,7 @@ jobs:
with:
key: ubuntu-22-cmake-hip
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Build with native CMake HIP support
id: cmake_build
@@ -599,6 +651,7 @@ jobs:
with:
key: ubuntu-22-cmake-musa
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Build with native CMake MUSA support
id: cmake_build
@@ -646,6 +699,7 @@ jobs:
with:
key: ubuntu-22-cmake-sycl
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Build
id: cmake_build
@@ -696,6 +750,7 @@ jobs:
with:
key: ubuntu-22-cmake-sycl-fp16
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Build
id: cmake_build
@@ -729,6 +784,7 @@ jobs:
with:
key: macOS-latest-cmake-ios
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Build
id: cmake_build
@@ -760,6 +816,7 @@ jobs:
with:
key: macOS-latest-cmake-tvos
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Build
id: cmake_build
@@ -821,6 +878,7 @@ jobs:
with:
key: macOS-latest-swift
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Download xcframework artifact
uses: actions/download-artifact@v4
@@ -863,6 +921,7 @@ jobs:
key: windows-msys2
variant: ccache
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Setup ${{ matrix.sys }}
uses: msys2/setup-msys2@v2
@@ -931,6 +990,7 @@ jobs:
key: windows-latest-cmake-${{ matrix.build }}
variant: ccache
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Download OpenBLAS
id: get_openblas
@@ -1035,8 +1095,10 @@ jobs:
with:
key: ubuntu-latest-cmake-cuda
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Build with CMake
+ # TODO: Remove GGML_CUDA_CUB_3DOT2 flag once CCCL 3.2 is bundled within CTK and that CTK version is used in this project
run: |
cmake -S . -B build -G Ninja \
-DLLAMA_CURL=OFF \
@@ -1046,7 +1108,8 @@ jobs:
-DCMAKE_CUDA_ARCHITECTURES=89-real \
-DCMAKE_EXE_LINKER_FLAGS=-Wl,--allow-shlib-undefined \
-DGGML_NATIVE=OFF \
- -DGGML_CUDA=ON
+ -DGGML_CUDA=ON \
+ -DGGML_CUDA_CUB_3DOT2=ON
cmake --build build
windows-2022-cmake-cuda:
@@ -1067,6 +1130,7 @@ jobs:
key: windows-cuda-${{ matrix.cuda }}
variant: ccache
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Install Cuda Toolkit
uses: ./.github/actions/windows-setup-cuda
@@ -1081,6 +1145,7 @@ jobs:
- name: Build
id: cmake_build
shell: cmd
+ # TODO: Remove GGML_CUDA_CUB_3DOT2 flag once CCCL 3.2 is bundled within CTK and that CTK version is used in this project
run: |
call "C:\Program Files\Microsoft Visual Studio\2022\Enterprise\VC\Auxiliary\Build\vcvarsall.bat" x64
cmake -S . -B build -G "Ninja Multi-Config" ^
@@ -1091,7 +1156,8 @@ jobs:
-DGGML_BACKEND_DL=ON ^
-DGGML_CPU_ALL_VARIANTS=ON ^
-DGGML_CUDA=ON ^
- -DGGML_RPC=ON
+ -DGGML_RPC=ON ^
+ -DGGML_CUDA_CUB_3DOT2=ON
set /A NINJA_JOBS=%NUMBER_OF_PROCESSORS%-1
cmake --build build --config Release -j %NINJA_JOBS% -t ggml
cmake --build build --config Release
@@ -1118,6 +1184,7 @@ jobs:
key: windows-latest-cmake-sycl
variant: ccache
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Install
run: |
@@ -1179,6 +1246,7 @@ jobs:
with:
key: ${{ github.job }}
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Build
id: cmake_build
@@ -1360,25 +1428,54 @@ jobs:
chip_type: ['910b', '310p']
build: ['Release']
runs-on: ${{ matrix.arch == 'aarch64' && 'ubuntu-24.04-arm' || 'ubuntu-24.04' }}
- container: ascendai/cann:${{ matrix.chip_type == '910b' && '8.3.rc1.alpha001-910b-openeuler22.03-py3.11' || '8.2.rc1-310p-openeuler22.03-py3.11' }}
steps:
- name: Checkout
uses: actions/checkout@v4
+ with:
+ fetch-depth: 0
- - name: Dependencies
+ - name: Free up disk space
+ uses: ggml-org/free-disk-space@v1.3.1
+ with:
+ tool-cache: true
+
+ - name: Set container image
+ id: cann-image
run: |
- yum update -y
- yum install -y git gcc gcc-c++ make cmake libcurl-devel
+ image="ascendai/cann:${{ matrix.chip_type == '910b' && '8.3.rc2-910b-openeuler24.03-py3.11' || '8.3.rc2-310p-openeuler24.03-py3.11' }}"
+ echo "image=${image}" >> "${GITHUB_OUTPUT}"
+
+ - name: Pull container image
+ run: docker pull "${{ steps.cann-image.outputs.image }}"
- name: Build
+ env:
+ BUILD_TYPE: ${{ matrix.build }}
+ SOC_TYPE: ascend${{ matrix.chip_type }}
run: |
- export LD_LIBRARY_PATH=${ASCEND_TOOLKIT_HOME}/lib64:${ASCEND_TOOLKIT_HOME}/$(uname -m)-linux/devlib/:${LD_LIBRARY_PATH}
+ HOST_UID=$(id -u)
+ HOST_GID=$(id -g)
- cmake -S . -B build \
- -DCMAKE_BUILD_TYPE=${{ matrix.build }} \
- -DGGML_CANN=on \
- -DSOC_TYPE=ascend${{ matrix.chip_type }}
- cmake --build build -j $(nproc)
+ docker run --rm \
+ -v "${PWD}:/workspace" \
+ -w /workspace \
+ -e SOC_TYPE=${SOC_TYPE} \
+ -e BUILD_TYPE=${BUILD_TYPE} \
+ "${{ steps.cann-image.outputs.image }}" \
+ bash -lc '
+ set -e
+ yum install -y --setopt=install_weak_deps=False --setopt=tsflags=nodocs git gcc gcc-c++ make cmake libcurl-devel
+ yum clean all && rm -rf /var/cache/yum
+ git config --global --add safe.directory "/workspace"
+ export LD_LIBRARY_PATH=${ASCEND_TOOLKIT_HOME}/lib64:${ASCEND_TOOLKIT_HOME}/$(uname -m)-linux/devlib/:${LD_LIBRARY_PATH}
+ cmake -S . -B build \
+ -DCMAKE_BUILD_TYPE=${BUILD_TYPE} \
+ -DGGML_CANN=on \
+ -DSOC_TYPE=${SOC_TYPE}
+ cmake --build build -j $(nproc)
+
+ chown -R '"${HOST_UID}"':'"${HOST_GID}"' /workspace/build
+ '
# TODO: simplify the following workflows using a matrix
# TODO: run lighter CI on PRs and the full CI only on master (if needed)
@@ -1395,6 +1492,7 @@ jobs:
with:
key: ggml-ci-x64-cpu-low-perf
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Dependencies
id: depends
@@ -1420,6 +1518,7 @@ jobs:
with:
key: ggml-ci-arm64-cpu-low-perf
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Dependencies
id: depends
@@ -1445,6 +1544,7 @@ jobs:
with:
key: ggml-ci-x64-cpu-high-perf
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Dependencies
id: depends
@@ -1470,6 +1570,7 @@ jobs:
with:
key: ggml-ci-arm64-cpu-high-perf
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Dependencies
id: depends
@@ -1495,6 +1596,7 @@ jobs:
with:
key: ggml-ci-arm64-cpu-high-perf-sve
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Dependencies
id: depends
@@ -1562,33 +1664,33 @@ jobs:
run: |
bash ./ci/run.sh ~/results/llama.cpp /mnt/llama.cpp
- ggml-ci-x64-amd-vulkan:
- runs-on: [self-hosted, Linux, X64, AMD]
+ # ggml-ci-x64-amd-vulkan:
+ # runs-on: [self-hosted, Linux, X64, AMD]
- steps:
- - name: Clone
- id: checkout
- uses: actions/checkout@v4
+ # steps:
+ # - name: Clone
+ # id: checkout
+ # uses: actions/checkout@v4
- - name: Test
- id: ggml-ci
- run: |
- vulkaninfo --summary
- GG_BUILD_VULKAN=1 bash ./ci/run.sh ~/results/llama.cpp /mnt/llama.cpp
+ # - name: Test
+ # id: ggml-ci
+ # run: |
+ # vulkaninfo --summary
+ # GG_BUILD_VULKAN=1 bash ./ci/run.sh ~/results/llama.cpp /mnt/llama.cpp
- ggml-ci-x64-amd-rocm:
- runs-on: [self-hosted, Linux, X64, AMD]
+ # ggml-ci-x64-amd-rocm:
+ # runs-on: [self-hosted, Linux, X64, AMD]
- steps:
- - name: Clone
- id: checkout
- uses: actions/checkout@v4
+ # steps:
+ # - name: Clone
+ # id: checkout
+ # uses: actions/checkout@v4
- - name: Test
- id: ggml-ci
- run: |
- amd-smi static
- GG_BUILD_ROCM=1 GG_BUILD_AMDGPU_TARGETS="gfx1101" bash ./ci/run.sh ~/results/llama.cpp /mnt/llama.cpp
+ # - name: Test
+ # id: ggml-ci
+ # run: |
+ # amd-smi static
+ # GG_BUILD_ROCM=1 GG_BUILD_AMDGPU_TARGETS="gfx1101" bash ./ci/run.sh ~/results/llama.cpp /mnt/llama.cpp
ggml-ci-mac-metal:
runs-on: [self-hosted, macOS, ARM64]
@@ -1630,6 +1732,7 @@ jobs:
with:
key: ggml-ci-arm64-cpu-kleidiai
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Dependencies
id: depends
@@ -1642,6 +1745,345 @@ jobs:
run: |
GG_BUILD_KLEIDIAI=1 GG_BUILD_EXTRA_TESTS_0=1 bash ./ci/run.sh ./tmp/results ./tmp/mnt
+ ubuntu-cpu-cmake-riscv64-native:
+ runs-on: RISCV64
+
+ steps:
+ - name: Install dependencies
+ run: |
+ sudo apt-get update
+
+ # Install necessary packages
+ sudo apt-get install -y libatomic1 libtsan2 gcc-14 g++-14 rustup cmake build-essential libssl-dev wget ccache git-lfs
+
+ # Set gcc-14 and g++-14 as the default compilers
+ sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-14 100
+ sudo update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-14 100
+ sudo ln -sf /usr/bin/gcc-14 /usr/bin/gcc
+ sudo ln -sf /usr/bin/g++-14 /usr/bin/g++
+
+ # Install Rust stable version
+ rustup install stable
+ rustup default stable
+
+ git lfs install
+
+ - name: Clone
+ id: checkout
+ uses: actions/checkout@v4
+
+ - name: Check environment
+ run: |
+ uname -a
+ gcc --version
+ g++ --version
+ ldd --version
+ cmake --version
+ rustc --version
+
+ - name: Setup ccache
+ run: |
+ # Set unique cache directory for this job
+ export CCACHE_DIR="$HOME/.ccache/cpu-cmake-rv64-native"
+ mkdir -p "$CCACHE_DIR"
+
+ # Configure ccache for optimal performance
+ ccache --set-config=max_size=5G
+ ccache --set-config=compression=true
+ ccache --set-config=compression_level=6
+ ccache --set-config=cache_dir="$CCACHE_DIR"
+
+ # Enable more aggressive caching
+ ccache --set-config=sloppiness=file_macro,time_macros,include_file_mtime,include_file_ctime
+ ccache --set-config=hash_dir=false
+
+ # Export for subsequent steps
+ echo "CCACHE_DIR=$CCACHE_DIR" >> $GITHUB_ENV
+ echo "PATH=/usr/lib/ccache:$PATH" >> $GITHUB_ENV
+
+ - name: Build
+ id: cmake_build
+ run: |
+ cmake -B build \
+ -DLLAMA_CURL=OFF \
+ -DLLAMA_OPENSSL=ON \
+ -DCMAKE_BUILD_TYPE=Release \
+ -DGGML_OPENMP=OFF \
+ -DLLAMA_BUILD_EXAMPLES=ON \
+ -DLLAMA_BUILD_TOOLS=ON \
+ -DLLAMA_BUILD_TESTS=ON \
+ -DCMAKE_C_COMPILER_LAUNCHER=ccache \
+ -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
+ -DGGML_RPC=ON \
+ -DCMAKE_C_COMPILER=riscv64-linux-gnu-gcc-14 \
+ -DCMAKE_CXX_COMPILER=riscv64-linux-gnu-g++-14
+
+ cmake --build build --config Release -j $(nproc)
+
+ - name: Test
+ id: cmake_test
+ run: |
+ cd build
+ ctest -L 'main|curl' --verbose --timeout 900
+
+ - name: Test llama2c conversion
+ id: llama2c_test
+ run: |
+ cd build
+ echo "Fetch tokenizer"
+ wget https://huggingface.co/karpathy/tinyllamas/resolve/main/stories260K/tok512.bin
+ echo "Fetch llama2c model"
+ wget https://huggingface.co/karpathy/tinyllamas/resolve/main/stories260K/stories260K.bin
+ ./bin/llama-convert-llama2c-to-ggml --copy-vocab-from-model ./tok512.bin --llama2c-model stories260K.bin --llama2c-output-model stories260K.gguf
+ ./bin/llama-completion -m stories260K.gguf -p "One day, Lily met a Shoggoth" -n 500 -c 256
+
+ ubuntu-cmake-sanitizer-riscv64-native:
+ runs-on: RISCV64
+
+ continue-on-error: true
+
+ strategy:
+ matrix:
+ sanitizer: [ADDRESS, THREAD, UNDEFINED]
+ build_type: [Debug]
+
+ steps:
+ - name: Install dependencies
+ run: |
+ sudo apt-get update
+
+ # Install necessary packages
+ sudo apt-get install -y libatomic1 libtsan2 gcc-14 g++-14 rustup cmake build-essential wget ccache git-lfs
+
+ # Set gcc-14 and g++-14 as the default compilers
+ sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-14 100
+ sudo update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-14 100
+ sudo ln -sf /usr/bin/gcc-14 /usr/bin/gcc
+ sudo ln -sf /usr/bin/g++-14 /usr/bin/g++
+
+ # Install Rust stable version
+ rustup install stable
+ rustup default stable
+
+ git lfs install
+
+ - name: GCC version check
+ run: |
+ gcc --version
+ g++ --version
+
+ - name: Clone
+ id: checkout
+ uses: actions/checkout@v4
+
+ - name: Setup ccache
+ run: |
+ # Unique cache directory per matrix combination
+ export CCACHE_DIR="$HOME/.ccache/sanitizer-${{ matrix.sanitizer }}-${{ matrix.build_type }}"
+ mkdir -p "$CCACHE_DIR"
+
+ # Configure ccache
+ ccache --set-config=max_size=5G
+ ccache --set-config=compression=true
+ ccache --set-config=compression_level=6
+ ccache --set-config=cache_dir="$CCACHE_DIR"
+ ccache --set-config=sloppiness=file_macro,time_macros,include_file_mtime,include_file_ctime
+ ccache --set-config=hash_dir=false
+
+ # Export for subsequent steps
+ echo "CCACHE_DIR=$CCACHE_DIR" >> $GITHUB_ENV
+ echo "PATH=/usr/lib/ccache:$PATH" >> $GITHUB_ENV
+
+ - name: Build
+ id: cmake_build
+ if: ${{ matrix.sanitizer != 'THREAD' }}
+ run: |
+ cmake -B build \
+ -DLLAMA_CURL=OFF \
+ -DCMAKE_BUILD_TYPE=${{ matrix.build_type }} \
+ -DGGML_OPENMP=ON \
+ -DLLAMA_BUILD_EXAMPLES=ON \
+ -DLLAMA_BUILD_TOOLS=ON \
+ -DLLAMA_BUILD_TESTS=OFF \
+ -DCMAKE_C_COMPILER_LAUNCHER=ccache \
+ -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
+ -DLLAMA_SANITIZE_${{ matrix.sanitizer }}=ON \
+ -DCMAKE_C_COMPILER=riscv64-linux-gnu-gcc-14 \
+ -DCMAKE_CXX_COMPILER=riscv64-linux-gnu-g++-14
+
+ cmake --build build --config ${{ matrix.build_type }} -j $(nproc)
+
+ - name: Build (no OpenMP)
+ id: cmake_build_no_openmp
+ if: ${{ matrix.sanitizer == 'THREAD' }}
+ run: |
+ cmake -B build \
+ -DLLAMA_CURL=OFF \
+ -DCMAKE_BUILD_TYPE=${{ matrix.build_type }} \
+ -DGGML_OPENMP=OFF \
+ -DLLAMA_BUILD_EXAMPLES=ON \
+ -DLLAMA_BUILD_TOOLS=ON \
+ -DLLAMA_BUILD_TESTS=OFF \
+ -DCMAKE_C_COMPILER_LAUNCHER=ccache \
+ -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
+ -DLLAMA_SANITIZE_${{ matrix.sanitizer }}=ON \
+ -DCMAKE_C_COMPILER=riscv64-linux-gnu-gcc-14 \
+ -DCMAKE_CXX_COMPILER=riscv64-linux-gnu-g++-14
+
+ cmake --build build --config ${{ matrix.build_type }} -j $(nproc)
+
+ - name: Test
+ id: cmake_test
+ run: |
+ cd build
+ ctest -L main --verbose --timeout 900
+
+
+ ubuntu-llguidance-riscv64-native:
+ runs-on: RISCV64
+ steps:
+ - name: Install dependencies
+ run: |
+ sudo apt-get update
+
+ # Install necessary packages
+ sudo apt-get install -y libatomic1 libtsan2 gcc-14 g++-14 rustup cmake build-essential wget ccache git-lfs
+
+ # Set gcc-14 and g++-14 as the default compilers
+ sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-14 100
+ sudo update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-14 100
+ sudo ln -sf /usr/bin/gcc-14 /usr/bin/gcc
+ sudo ln -sf /usr/bin/g++-14 /usr/bin/g++
+
+ # Install Rust stable version
+ rustup install stable
+ rustup default stable
+
+ git lfs install
+
+ - name: GCC version check
+ run: |
+ gcc --version
+ g++ --version
+
+ - name: Clone
+ id: checkout
+ uses: actions/checkout@v4
+
+ - name: Setup ccache
+ run: |
+ export CCACHE_DIR="$HOME/.ccache/llguidance-riscv64"
+ mkdir -p "$CCACHE_DIR"
+
+ ccache --set-config=max_size=5G
+ ccache --set-config=compression=true
+ ccache --set-config=compression_level=6
+ ccache --set-config=cache_dir="$CCACHE_DIR"
+ ccache --set-config=sloppiness=file_macro,time_macros,include_file_mtime,include_file_ctime
+ ccache --set-config=hash_dir=false
+
+ echo "CCACHE_DIR=$CCACHE_DIR" >> $GITHUB_ENV
+ echo "PATH=/usr/lib/ccache:$PATH" >> $GITHUB_ENV
+
+ - name: Build
+ id: cmake_build
+ run: |
+ cmake -B build \
+ -DLLAMA_CURL=OFF \
+ -DCMAKE_BUILD_TYPE=Release \
+ -DGGML_OPENMP=OFF \
+ -DLLAMA_BUILD_EXAMPLES=ON \
+ -DLLAMA_BUILD_TOOLS=ON \
+ -DLLAMA_BUILD_TESTS=OFF \
+ -DCMAKE_C_COMPILER_LAUNCHER=ccache \
+ -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
+ -DLLAMA_LLGUIDANCE=ON \
+ -DCMAKE_C_COMPILER=riscv64-linux-gnu-gcc-14 \
+ -DCMAKE_CXX_COMPILER=riscv64-linux-gnu-g++-14
+
+ cmake --build build --config Release -j $(nproc)
+
+ - name: Test
+ id: cmake_test
+ run: |
+ cd build
+ ctest -L main --verbose --timeout 900
+
+
+ ubuntu-cmake-rpc-riscv64-native:
+ runs-on: RISCV64
+
+ continue-on-error: true
+
+ steps:
+ - name: Install dependencies
+ run: |
+ sudo apt-get update
+
+ # Install necessary packages
+ sudo apt-get install -y libatomic1 libtsan2 gcc-14 g++-14 rustup cmake build-essential libssl-dev wget ccache git-lfs
+
+ # Set gcc-14 and g++-14 as the default compilers
+ sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-14 100
+ sudo update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-14 100
+ sudo ln -sf /usr/bin/gcc-14 /usr/bin/gcc
+ sudo ln -sf /usr/bin/g++-14 /usr/bin/g++
+
+ # Install Rust stable version
+ rustup install stable
+ rustup default stable
+
+ git lfs install
+
+ - name: GCC version check
+ run: |
+ gcc --version
+ g++ --version
+
+ - name: Clone
+ id: checkout
+ uses: actions/checkout@v4
+
+ - name: Setup ccache
+ run: |
+ export CCACHE_DIR="$HOME/.ccache/rpc-riscv64"
+ mkdir -p "$CCACHE_DIR"
+
+ ccache --set-config=max_size=5G
+ ccache --set-config=compression=true
+ ccache --set-config=compression_level=6
+ ccache --set-config=cache_dir="$CCACHE_DIR"
+ ccache --set-config=sloppiness=file_macro,time_macros,include_file_mtime,include_file_ctime
+ ccache --set-config=hash_dir=false
+
+ echo "CCACHE_DIR=$CCACHE_DIR" >> $GITHUB_ENV
+ echo "PATH=/usr/lib/ccache:$PATH" >> $GITHUB_ENV
+
+ - name: Build
+ id: cmake_build
+ run: |
+ cmake -B build \
+ -DLLAMA_CURL=OFF \
+ -DLLAMA_OPENSSL=ON \
+ -DCMAKE_BUILD_TYPE=Release \
+ -DGGML_OPENMP=OFF \
+ -DLLAMA_BUILD_EXAMPLES=ON \
+ -DLLAMA_BUILD_TOOLS=ON \
+ -DLLAMA_BUILD_TESTS=ON \
+ -DCMAKE_C_COMPILER_LAUNCHER=ccache \
+ -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
+ -DCMAKE_C_COMPILER=riscv64-linux-gnu-gcc-14 \
+ -DCMAKE_CXX_COMPILER=riscv64-linux-gnu-g++-14 \
+ -DGGML_RPC=ON
+
+ cmake --build build --config Release -j $(nproc)
+
+ - name: Test
+ id: cmake_test
+ run: |
+ cd build
+ ctest -L main --verbose
+
ggml-ci-arm64-graviton4-kleidiai:
runs-on: ah-ubuntu_22_04-c8g_8x
@@ -1682,6 +2124,7 @@ jobs:
with:
key: ggml-ci-arm64-graviton4-kleidiai
evict-old-files: 1d
+ save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
- name: Test
id: ggml-ci
diff --git a/.github/workflows/docker.yml b/.github/workflows/docker.yml
index 7ca11b1dff..d9fe0686d3 100644
--- a/.github/workflows/docker.yml
+++ b/.github/workflows/docker.yml
@@ -40,13 +40,13 @@ jobs:
# https://github.com/ggml-org/llama.cpp/issues/11888
#- { tag: "cpu", dockerfile: ".devops/cpu.Dockerfile", platforms: "linux/amd64,linux/arm64", full: true, light: true, server: true, free_disk_space: false }
- { tag: "cpu", dockerfile: ".devops/cpu.Dockerfile", platforms: "linux/amd64", full: true, light: true, server: true, free_disk_space: false, runs_on: "ubuntu-22.04" }
- - { tag: "cuda", dockerfile: ".devops/cuda.Dockerfile", platforms: "linux/amd64", full: true, light: true, server: true, free_disk_space: true, runs_on: "ubuntu-22.04" }
+ - { tag: "cuda cuda12", dockerfile: ".devops/cuda.Dockerfile", platforms: "linux/amd64", full: true, light: true, server: true, free_disk_space: true, runs_on: "ubuntu-22.04", cuda_version: "12.4.0", ubuntu_version: "22.04" }
+ - { tag: "cuda13", dockerfile: ".devops/cuda-new.Dockerfile", platforms: "linux/amd64", full: true, light: true, server: true, free_disk_space: true, runs_on: "ubuntu-22.04", cuda_version: "13.1.0", ubuntu_version: "24.04" }
- { tag: "musa", dockerfile: ".devops/musa.Dockerfile", platforms: "linux/amd64", full: true, light: true, server: true, free_disk_space: true, runs_on: "ubuntu-22.04" }
- { tag: "intel", dockerfile: ".devops/intel.Dockerfile", platforms: "linux/amd64", full: true, light: true, server: true, free_disk_space: true, runs_on: "ubuntu-22.04" }
- { tag: "vulkan", dockerfile: ".devops/vulkan.Dockerfile", platforms: "linux/amd64", full: true, light: true, server: true, free_disk_space: false, runs_on: "ubuntu-22.04" }
- { tag: "s390x", dockerfile: ".devops/s390x.Dockerfile", platforms: "linux/s390x", full: true, light: true, server: true, free_disk_space: false, runs_on: "ubuntu-22.04-s390x" }
- # Note: the rocm images are failing due to a compiler error and are disabled until this is fixed to allow the workflow to complete
- #- {tag: "rocm", dockerfile: ".devops/rocm.Dockerfile", platforms: "linux/amd64,linux/arm64", full: true, light: true, server: true, free_disk_space: true }
+ - { tag: "rocm", dockerfile: ".devops/rocm.Dockerfile", platforms: "linux/amd64", full: true, light: true, server: true, free_disk_space: true, runs_on: "ubuntu-22.04" }
steps:
- name: Check out the repo
uses: actions/checkout@v4
@@ -81,18 +81,21 @@ jobs:
run: |
REPO_OWNER="${GITHUB_REPOSITORY_OWNER@L}" # to lower case
REPO_NAME="${{ github.event.repository.name }}"
+ PREFIX="ghcr.io/${REPO_OWNER}/${REPO_NAME}:"
# list all tags possible
- if [[ "${{ matrix.config.tag }}" == "cpu" ]]; then
- TYPE=""
- else
- TYPE="-${{ matrix.config.tag }}"
- fi
- PREFIX="ghcr.io/${REPO_OWNER}/${REPO_NAME}:"
- CACHETAGS="${PREFIX}buildcache${TYPE}"
- FULLTAGS="${PREFIX}full${TYPE},${PREFIX}full${TYPE}-${{ steps.srctag.outputs.name }}"
- LIGHTTAGS="${PREFIX}light${TYPE},${PREFIX}light${TYPE}-${{ steps.srctag.outputs.name }}"
- SERVERTAGS="${PREFIX}server${TYPE},${PREFIX}server${TYPE}-${{ steps.srctag.outputs.name }}"
+ tags="${{ matrix.config.tag }}"
+ for tag in $tags; do
+ if [[ "$tag" == "cpu" ]]; then
+ TYPE=""
+ else
+ TYPE="-$tag"
+ fi
+ CACHETAGS="${PREFIX}buildcache${TYPE}"
+ FULLTAGS="${FULLTAGS:+$FULLTAGS,}${PREFIX}full${TYPE},${PREFIX}full${TYPE}-${{ steps.srctag.outputs.name }}"
+ LIGHTTAGS="${LIGHTTAGS:+$LIGHTTAGS,}${PREFIX}light${TYPE},${PREFIX}light${TYPE}-${{ steps.srctag.outputs.name }}"
+ SERVERTAGS="${SERVERTAGS:+$SERVERTAGS,}${PREFIX}server${TYPE},${PREFIX}server${TYPE}-${{ steps.srctag.outputs.name }}"
+ done
echo "cache_output_tags=$CACHETAGS" >> $GITHUB_OUTPUT
echo "full_output_tags=$FULLTAGS" >> $GITHUB_OUTPUT
echo "light_output_tags=$LIGHTTAGS" >> $GITHUB_OUTPUT
@@ -133,6 +136,9 @@ jobs:
file: ${{ matrix.config.dockerfile }}
target: full
provenance: false
+ build-args: |
+ ${{ matrix.config.ubuntu_version && format('UBUNTU_VERSION={0}', matrix.config.ubuntu_version) || '' }}
+ ${{ matrix.config.cuda_version && format('CUDA_VERSION={0}', matrix.config.cuda_version) || '' }}
# using github experimental cache
#cache-from: type=gha
#cache-to: type=gha,mode=max
@@ -155,6 +161,9 @@ jobs:
file: ${{ matrix.config.dockerfile }}
target: light
provenance: false
+ build-args: |
+ ${{ matrix.config.ubuntu_version && format('UBUNTU_VERSION={0}', matrix.config.ubuntu_version) || '' }}
+ ${{ matrix.config.cuda_version && format('CUDA_VERSION={0}', matrix.config.cuda_version) || '' }}
# using github experimental cache
#cache-from: type=gha
#cache-to: type=gha,mode=max
@@ -177,6 +186,9 @@ jobs:
file: ${{ matrix.config.dockerfile }}
target: server
provenance: false
+ build-args: |
+ ${{ matrix.config.ubuntu_version && format('UBUNTU_VERSION={0}', matrix.config.ubuntu_version) || '' }}
+ ${{ matrix.config.cuda_version && format('CUDA_VERSION={0}', matrix.config.cuda_version) || '' }}
# using github experimental cache
#cache-from: type=gha
#cache-to: type=gha,mode=max
diff --git a/.github/workflows/release.yml b/.github/workflows/release.yml
index 0d5739c24b..bf5ebb7559 100644
--- a/.github/workflows/release.yml
+++ b/.github/workflows/release.yml
@@ -66,13 +66,13 @@ jobs:
id: pack_artifacts
run: |
cp LICENSE ./build/bin/
- zip -r llama-${{ steps.tag.outputs.name }}-bin-macos-arm64.zip ./build/bin/*
+ tar -czvf llama-${{ steps.tag.outputs.name }}-bin-macos-arm64.tar.gz -s ",./,llama-${{ steps.tag.outputs.name }}/," -C ./build/bin .
- name: Upload artifacts
uses: actions/upload-artifact@v4
with:
- path: llama-${{ steps.tag.outputs.name }}-bin-macos-arm64.zip
- name: llama-bin-macos-arm64.zip
+ path: llama-${{ steps.tag.outputs.name }}-bin-macos-arm64.tar.gz
+ name: llama-bin-macos-arm64.tar.gz
macOS-x64:
runs-on: macos-15-intel
@@ -120,13 +120,13 @@ jobs:
id: pack_artifacts
run: |
cp LICENSE ./build/bin/
- zip -r llama-${{ steps.tag.outputs.name }}-bin-macos-x64.zip ./build/bin/*
+ tar -czvf llama-${{ steps.tag.outputs.name }}-bin-macos-x64.tar.gz -s ",./,llama-${{ steps.tag.outputs.name }}/," -C ./build/bin .
- name: Upload artifacts
uses: actions/upload-artifact@v4
with:
- path: llama-${{ steps.tag.outputs.name }}-bin-macos-x64.zip
- name: llama-bin-macos-x64.zip
+ path: llama-${{ steps.tag.outputs.name }}-bin-macos-x64.tar.gz
+ name: llama-bin-macos-x64.tar.gz
ubuntu-22-cpu:
strategy:
@@ -182,13 +182,13 @@ jobs:
id: pack_artifacts
run: |
cp LICENSE ./build/bin/
- zip -r llama-${{ steps.tag.outputs.name }}-bin-ubuntu-${{ matrix.build }}.zip ./build/bin/*
+ tar -czvf llama-${{ steps.tag.outputs.name }}-bin-ubuntu-${{ matrix.build }}.tar.gz --transform "s,./,llama-${{ steps.tag.outputs.name }}/," -C ./build/bin .
- name: Upload artifacts
uses: actions/upload-artifact@v4
with:
- path: llama-${{ steps.tag.outputs.name }}-bin-ubuntu-${{ matrix.build }}.zip
- name: llama-bin-ubuntu-${{ matrix.build }}.zip
+ path: llama-${{ steps.tag.outputs.name }}-bin-ubuntu-${{ matrix.build }}.tar.gz
+ name: llama-bin-ubuntu-${{ matrix.build }}.tar.gz
ubuntu-22-vulkan:
runs-on: ubuntu-22.04
@@ -235,13 +235,13 @@ jobs:
id: pack_artifacts
run: |
cp LICENSE ./build/bin/
- zip -r llama-${{ steps.tag.outputs.name }}-bin-ubuntu-vulkan-x64.zip ./build/bin/*
+ tar -czvf llama-${{ steps.tag.outputs.name }}-bin-ubuntu-vulkan-x64.tar.gz --transform "s,./,llama-${{ steps.tag.outputs.name }}/," -C ./build/bin .
- name: Upload artifacts
uses: actions/upload-artifact@v4
with:
- path: llama-${{ steps.tag.outputs.name }}-bin-ubuntu-vulkan-x64.zip
- name: llama-bin-ubuntu-vulkan-x64.zip
+ path: llama-${{ steps.tag.outputs.name }}-bin-ubuntu-vulkan-x64.tar.gz
+ name: llama-bin-ubuntu-vulkan-x64.tar.gz
windows-cpu:
runs-on: windows-2025
@@ -298,7 +298,7 @@ jobs:
run: |
Copy-Item $env:CURL_PATH\bin\libcurl-${{ matrix.arch }}.dll .\build\bin\Release\
Copy-Item "C:\Program Files\Microsoft Visual Studio\2022\Enterprise\VC\Redist\MSVC\14.44.35112\debug_nonredist\${{ matrix.arch }}\Microsoft.VC143.OpenMP.LLVM\libomp140.${{ matrix.arch == 'x64' && 'x86_64' || 'aarch64' }}.dll" .\build\bin\Release\
- 7z a llama-bin-win-cpu-${{ matrix.arch }}.zip .\build\bin\Release\*
+ 7z a -snl llama-bin-win-cpu-${{ matrix.arch }}.zip .\build\bin\Release\*
- name: Upload artifacts
uses: actions/upload-artifact@v4
@@ -380,7 +380,7 @@ jobs:
- name: Pack artifacts
id: pack_artifacts
run: |
- 7z a llama-bin-win-${{ matrix.backend }}-${{ matrix.arch }}.zip .\build\bin\Release\${{ matrix.target }}.dll
+ 7z a -snl llama-bin-win-${{ matrix.backend }}-${{ matrix.arch }}.zip .\build\bin\Release\${{ matrix.target }}.dll
- name: Upload artifacts
uses: actions/upload-artifact@v4
@@ -393,7 +393,7 @@ jobs:
strategy:
matrix:
- cuda: ['12.4']
+ cuda: ['12.4', '13.1']
steps:
- name: Clone
@@ -420,6 +420,7 @@ jobs:
- name: Build
id: cmake_build
shell: cmd
+ # TODO: Remove GGML_CUDA_CUB_3DOT2 flag once CCCL 3.2 is bundled within CTK and that CTK version is used in this project
run: |
call "C:\Program Files\Microsoft Visual Studio\2022\Enterprise\VC\Auxiliary\Build\vcvarsall.bat" x64
cmake -S . -B build -G "Ninja Multi-Config" ^
@@ -427,14 +428,15 @@ jobs:
-DGGML_NATIVE=OFF ^
-DGGML_CPU=OFF ^
-DGGML_CUDA=ON ^
- -DLLAMA_CURL=OFF
+ -DLLAMA_CURL=OFF ^
+ -DGGML_CUDA_CUB_3DOT2=ON
set /A NINJA_JOBS=%NUMBER_OF_PROCESSORS%-1
cmake --build build --config Release -j %NINJA_JOBS% --target ggml-cuda
- name: Pack artifacts
id: pack_artifacts
run: |
- 7z a llama-bin-win-cuda-${{ matrix.cuda }}-x64.zip .\build\bin\Release\ggml-cuda.dll
+ 7z a -snl llama-bin-win-cuda-${{ matrix.cuda }}-x64.zip .\build\bin\Release\ggml-cuda.dll
- name: Upload artifacts
uses: actions/upload-artifact@v4
@@ -448,6 +450,7 @@ jobs:
$dst='.\build\bin\cudart\'
robocopy "${{env.CUDA_PATH}}\bin" $dst cudart64_*.dll cublas64_*.dll cublasLt64_*.dll
robocopy "${{env.CUDA_PATH}}\lib" $dst cudart64_*.dll cublas64_*.dll cublasLt64_*.dll
+ robocopy "${{env.CUDA_PATH}}\bin\x64" $dst cudart64_*.dll cublas64_*.dll cublasLt64_*.dll
7z a cudart-llama-bin-win-cuda-${{ matrix.cuda }}-x64.zip $dst\*
- name: Upload Cuda runtime
@@ -517,6 +520,8 @@ jobs:
cp "${{ env.ONEAPI_ROOT }}/compiler/latest/bin/libmmd.dll" ./build/bin
cp "${{ env.ONEAPI_ROOT }}/compiler/latest/bin/libiomp5md.dll" ./build/bin
cp "${{ env.ONEAPI_ROOT }}/compiler/latest/bin/sycl-ls.exe" ./build/bin
+ cp "${{ env.ONEAPI_ROOT }}/compiler/latest/bin/libsycl-fallback-bfloat16.spv" ./build/bin
+ cp "${{ env.ONEAPI_ROOT }}/compiler/latest/bin/libsycl-native-bfloat16.spv" ./build/bin
cp "${{ env.ONEAPI_ROOT }}/dnnl/latest/bin/dnnl.dll" ./build/bin
cp "${{ env.ONEAPI_ROOT }}/tbb/latest/bin/tbb12.dll" ./build/bin
@@ -526,7 +531,7 @@ jobs:
cp "${{ env.ONEAPI_ROOT }}/umf/latest/bin/umf.dll" ./build/bin
echo "cp oneAPI running time dll files to ./build/bin done"
- 7z a llama-bin-win-sycl-x64.zip ./build/bin/*
+ 7z a -snl llama-bin-win-sycl-x64.zip ./build/bin/*
- name: Upload the release package
uses: actions/upload-artifact@v4
@@ -632,7 +637,7 @@ jobs:
- name: Pack artifacts
id: pack_artifacts
run: |
- 7z a llama-bin-win-hip-${{ matrix.name }}-x64.zip .\build\bin\*
+ 7z a -snl llama-bin-win-hip-${{ matrix.name }}-x64.zip .\build\bin\*
- name: Upload artifacts
uses: actions/upload-artifact@v4
@@ -685,13 +690,16 @@ jobs:
- name: Pack artifacts
id: pack_artifacts
run: |
- zip --symlinks -r llama-${{ steps.tag.outputs.name }}-xcframework.zip build-apple/llama.xcframework
+ # Zip file is required for Swift Package Manager, which does not support tar.gz for binary targets.
+ # For more details, see https://developer.apple.com/documentation/xcode/distributing-binary-frameworks-as-swift-packages
+ zip -r -y llama-${{ steps.tag.outputs.name }}-xcframework.zip build-apple/llama.xcframework
- name: Upload artifacts
uses: actions/upload-artifact@v4
with:
path: llama-${{ steps.tag.outputs.name }}-xcframework.zip
- name: llama-${{ steps.tag.outputs.name }}-xcframework
+ name: llama-${{ steps.tag.outputs.name }}-xcframework.zip
+
openEuler-cann:
strategy:
@@ -700,28 +708,54 @@ jobs:
chip_type: ['910b', '310p']
build: ['Release']
runs-on: ${{ matrix.arch == 'aarch64' && 'ubuntu-24.04-arm' || 'ubuntu-24.04' }}
- container: ascendai/cann:${{ matrix.chip_type == '910b' && '8.3.rc1.alpha001-910b-openeuler22.03-py3.11' || '8.2.rc1-310p-openeuler22.03-py3.11' }}
steps:
- name: Checkout
uses: actions/checkout@v4
with:
fetch-depth: 0
- - name: Dependencies
+ - name: Free up disk space
+ uses: ggml-org/free-disk-space@v1.3.1
+ with:
+ tool-cache: true
+
+ - name: Set container image
+ id: cann-image
run: |
- yum update -y
- yum install -y git gcc gcc-c++ make cmake libcurl-devel
- git config --global --add safe.directory "$GITHUB_WORKSPACE"
+ image="ascendai/cann:${{ matrix.chip_type == '910b' && '8.3.rc2-910b-openeuler24.03-py3.11' || '8.3.rc2-310p-openeuler24.03-py3.11' }}"
+ echo "image=${image}" >> "${GITHUB_OUTPUT}"
+
+ - name: Pull container image
+ run: docker pull "${{ steps.cann-image.outputs.image }}"
- name: Build
+ env:
+ BUILD_TYPE: ${{ matrix.build }}
+ SOC_TYPE: ascend${{ matrix.chip_type }}
run: |
- export LD_LIBRARY_PATH=${ASCEND_TOOLKIT_HOME}/lib64:${ASCEND_TOOLKIT_HOME}/$(uname -m)-linux/devlib/:${LD_LIBRARY_PATH}
+ HOST_UID=$(id -u)
+ HOST_GID=$(id -g)
- cmake -S . -B build \
- -DCMAKE_BUILD_TYPE=${{ matrix.build }} \
- -DGGML_CANN=on \
- -DSOC_TYPE=ascend${{ matrix.chip_type }}
- cmake --build build -j $(nproc)
+ docker run --rm \
+ -v "${PWD}:/workspace" \
+ -w /workspace \
+ -e SOC_TYPE=${SOC_TYPE} \
+ -e BUILD_TYPE=${BUILD_TYPE} \
+ "${{ steps.cann-image.outputs.image }}" \
+ bash -lc '
+ set -e
+ yum install -y --setopt=install_weak_deps=False --setopt=tsflags=nodocs git gcc gcc-c++ make cmake libcurl-devel
+ yum clean all && rm -rf /var/cache/yum
+ git config --global --add safe.directory "/workspace"
+ export LD_LIBRARY_PATH=${ASCEND_TOOLKIT_HOME}/lib64:${ASCEND_TOOLKIT_HOME}/$(uname -m)-linux/devlib/:${LD_LIBRARY_PATH}
+ cmake -S . -B build \
+ -DCMAKE_BUILD_TYPE=${BUILD_TYPE} \
+ -DGGML_CANN=on \
+ -DSOC_TYPE=${SOC_TYPE}
+ cmake --build build -j $(nproc)
+
+ chown -R '"${HOST_UID}"':'"${HOST_GID}"' /workspace/build
+ '
- name: Determine tag name
id: tag
@@ -730,13 +764,13 @@ jobs:
- name: Pack artifacts
run: |
cp LICENSE ./build/bin/
- zip -r llama-${{ steps.tag.outputs.name }}-bin-${{ matrix.chip_type }}-openEuler-${{ matrix.arch }}.zip ./build/bin/*
+ tar -czvf llama-${{ steps.tag.outputs.name }}-bin-${{ matrix.chip_type }}-openEuler-${{ matrix.arch }}.tar.gz --transform "s,./,llama-${{ steps.tag.outputs.name }}/," -C ./build/bin .
- name: Upload artifacts
uses: actions/upload-artifact@v4
with:
- path: llama-${{ steps.tag.outputs.name }}-bin-${{ matrix.chip_type }}-openEuler-${{ matrix.arch }}.zip
- name: llama-bin-${{ matrix.chip_type }}-openEuler-${{ matrix.arch }}.zip
+ path: llama-${{ steps.tag.outputs.name }}-bin-${{ matrix.chip_type }}-openEuler-${{ matrix.arch }}.tar.gz
+ name: llama-bin-${{ matrix.chip_type }}-openEuler-${{ matrix.arch }}.tar.gz
release:
if: ${{ ( github.event_name == 'push' && github.ref == 'refs/heads/master' ) || github.event.inputs.create_release == 'true' }}
@@ -814,6 +848,7 @@ jobs:
echo "Moving other artifacts..."
mv -v artifact/*.zip release
+ mv -v artifact/*.tar.gz release
- name: Create release
id: create_release
@@ -822,6 +857,37 @@ jobs:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
with:
tag_name: ${{ steps.tag.outputs.name }}
+ body: |
+
+
+ ${{ github.event.head_commit.message }}
+
+
+
+ **macOS/iOS:**
+ - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-bin-macos-arm64.tar.gz)
+ - [macOS Intel (x64)](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-bin-macos-x64.tar.gz)
+ - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-xcframework.zip)
+
+ **Linux:**
+ - [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-bin-ubuntu-x64.tar.gz)
+ - [Ubuntu x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-bin-ubuntu-vulkan-x64.tar.gz)
+ - [Ubuntu s390x (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-bin-ubuntu-s390x.tar.gz)
+
+ **Windows:**
+ - [Windows x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-bin-win-cpu-x64.zip)
+ - [Windows arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-bin-win-cpu-arm64.zip)
+ - [Windows x64 (CUDA 12)](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-bin-win-cuda-12.4-x64.zip) - [CUDA 12.4 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/cudart-llama-bin-win-cuda-12.4-x64.zip)
+ - [Windows x64 (CUDA 13)](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-bin-win-cuda-13.1-x64.zip) - [CUDA 13.1 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/cudart-llama-bin-win-cuda-13.1-x64.zip)
+ - [Windows x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-bin-win-vulkan-x64.zip)
+ - [Windows x64 (SYCL)](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-bin-win-sycl-x64.zip)
+ - [Windows x64 (HIP)](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-bin-win-hip-radeon-x64.zip)
+
+ **openEuler:**
+ - [openEuler x86 (310p)](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-bin-310p-openEuler-x86.tar.gz)
+ - [openEuler x86 (910b)](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-bin-910b-openEuler-x86.tar.gz)
+ - [openEuler aarch64 (310p)](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-bin-310p-openEuler-aarch64.tar.gz)
+ - [openEuler aarch64 (910b)](https://github.com/ggml-org/llama.cpp/releases/download/${{ steps.tag.outputs.name }}/llama-${{ steps.tag.outputs.name }}-bin-910b-openEuler-aarch64.tar.gz)
- name: Upload release
id: upload_release
@@ -833,7 +899,7 @@ jobs:
const fs = require('fs');
const release_id = '${{ steps.create_release.outputs.id }}';
for (let file of await fs.readdirSync('./release')) {
- if (path.extname(file) === '.zip') {
+ if (path.extname(file) === '.zip' || file.endsWith('.tar.gz')) {
console.log('uploadReleaseAsset', file);
await github.repos.uploadReleaseAsset({
owner: context.repo.owner,
diff --git a/.github/workflows/server-webui.yml b/.github/workflows/server-webui.yml
new file mode 100644
index 0000000000..544c4ad408
--- /dev/null
+++ b/.github/workflows/server-webui.yml
@@ -0,0 +1,225 @@
+# Server WebUI build and tests
+name: Server WebUI
+
+on:
+ workflow_dispatch: # allows manual triggering
+ inputs:
+ sha:
+ description: 'Commit SHA1 to build'
+ required: false
+ type: string
+ slow_tests:
+ description: 'Run slow tests'
+ required: true
+ type: boolean
+ push:
+ branches:
+ - master
+ paths: ['.github/workflows/server-webui.yml', 'tools/server/webui/**.*', 'tools/server/tests/**.*', 'tools/server/public/**']
+ pull_request:
+ types: [opened, synchronize, reopened]
+ paths: ['.github/workflows/server-webui.yml', 'tools/server/webui/**.*', 'tools/server/tests/**.*', 'tools/server/public/**']
+
+env:
+ LLAMA_LOG_COLORS: 1
+ LLAMA_LOG_PREFIX: 1
+ LLAMA_LOG_TIMESTAMPS: 1
+ LLAMA_LOG_VERBOSITY: 10
+
+concurrency:
+ group: ${{ github.workflow }}-${{ github.ref }}-${{ github.head_ref || github.run_id }}
+ cancel-in-progress: true
+
+jobs:
+ webui-check:
+ name: WebUI Checks
+ runs-on: ubuntu-latest
+ continue-on-error: true
+ steps:
+ - name: Checkout code
+ uses: actions/checkout@v4
+ with:
+ fetch-depth: 0
+ ref: ${{ github.event.inputs.sha || github.event.pull_request.head.sha || github.sha || github.head_ref || github.ref_name }}
+
+ - name: Setup Node.js
+ id: node
+ uses: actions/setup-node@v4
+ with:
+ node-version: "22"
+ cache: "npm"
+ cache-dependency-path: "tools/server/webui/package-lock.json"
+
+ - name: Install dependencies
+ id: setup
+ if: ${{ steps.node.conclusion == 'success' }}
+ run: npm ci
+ working-directory: tools/server/webui
+
+ - name: Run type checking
+ if: ${{ always() && steps.setup.conclusion == 'success' }}
+ run: npm run check
+ working-directory: tools/server/webui
+
+ - name: Run linting
+ if: ${{ always() && steps.setup.conclusion == 'success' }}
+ run: npm run lint
+ working-directory: tools/server/webui
+
+ - name: Build application
+ if: ${{ always() && steps.setup.conclusion == 'success' }}
+ run: npm run build
+ working-directory: tools/server/webui
+
+ - name: Install Playwright browsers
+ id: playwright
+ if: ${{ always() && steps.setup.conclusion == 'success' }}
+ run: npx playwright install --with-deps
+ working-directory: tools/server/webui
+
+ - name: Build Storybook
+ if: ${{ always() && steps.playwright.conclusion == 'success' }}
+ run: npm run build-storybook
+ working-directory: tools/server/webui
+
+ - name: Run Client tests
+ if: ${{ always() && steps.playwright.conclusion == 'success' }}
+ run: npm run test:client
+ working-directory: tools/server/webui
+
+ - name: Run Unit tests
+ if: ${{ always() && steps.playwright.conclusion == 'success' }}
+ run: npm run test:unit
+ working-directory: tools/server/webui
+
+ - name: Run UI tests
+ if: ${{ always() && steps.playwright.conclusion == 'success' }}
+ run: npm run test:ui -- --testTimeout=60000
+ working-directory: tools/server/webui
+
+ - name: Run E2E tests
+ if: ${{ always() && steps.playwright.conclusion == 'success' }}
+ run: npm run test:e2e
+ working-directory: tools/server/webui
+
+ server-build:
+ runs-on: ubuntu-latest
+
+ strategy:
+ matrix:
+ sanitizer: [ADDRESS, UNDEFINED] # THREAD is broken
+ build_type: [RelWithDebInfo]
+ include:
+ - build_type: Release
+ sanitizer: ""
+ fail-fast: false # While -DLLAMA_SANITIZE_THREAD=ON is broken
+
+ steps:
+ - name: Dependencies
+ id: depends
+ run: |
+ sudo apt-get update
+ sudo apt-get -y install \
+ build-essential \
+ xxd \
+ git \
+ cmake \
+ curl \
+ wget \
+ language-pack-en \
+ libssl-dev
+
+ - name: Clone
+ id: checkout
+ uses: actions/checkout@v4
+ with:
+ fetch-depth: 0
+ ref: ${{ github.event.inputs.sha || github.event.pull_request.head.sha || github.sha || github.head_ref || github.ref_name }}
+
+ - name: Python setup
+ id: setup_python
+ uses: actions/setup-python@v5
+ with:
+ python-version: '3.11'
+
+ - name: Tests dependencies
+ id: test_dependencies
+ run: |
+ pip install -r tools/server/tests/requirements.txt
+
+ - name: Setup Node.js for WebUI
+ uses: actions/setup-node@v4
+ with:
+ node-version: "22"
+ cache: "npm"
+ cache-dependency-path: "tools/server/webui/package-lock.json"
+
+ - name: Install WebUI dependencies
+ run: npm ci
+ working-directory: tools/server/webui
+
+ - name: Build WebUI
+ run: npm run build
+ working-directory: tools/server/webui
+
+ - name: Build (no OpenMP)
+ id: cmake_build_no_openmp
+ if: ${{ matrix.sanitizer == 'THREAD' }}
+ run: |
+ cmake -B build \
+ -DGGML_NATIVE=OFF \
+ -DLLAMA_CURL=OFF \
+ -DLLAMA_OPENSSL=ON \
+ -DLLAMA_BUILD_SERVER=ON \
+ -DCMAKE_BUILD_TYPE=${{ matrix.build_type }} \
+ -DLLAMA_SANITIZE_${{ matrix.sanitizer }}=ON \
+ -DGGML_OPENMP=OFF ;
+ cmake --build build --config ${{ matrix.build_type }} -j $(nproc) --target llama-server
+
+ - name: Build (sanitizers)
+ id: cmake_build_sanitizers
+ if: ${{ matrix.sanitizer != '' && matrix.sanitizer != 'THREAD' }}
+ run: |
+ cmake -B build \
+ -DGGML_NATIVE=OFF \
+ -DLLAMA_CURL=OFF \
+ -DLLAMA_OPENSSL=ON \
+ -DLLAMA_BUILD_SERVER=ON \
+ -DCMAKE_BUILD_TYPE=${{ matrix.build_type }} \
+ -DLLAMA_SANITIZE_${{ matrix.sanitizer }}=ON ;
+ cmake --build build --config ${{ matrix.build_type }} -j $(nproc) --target llama-server
+
+ - name: Build (sanitizers)
+ id: cmake_build
+ if: ${{ matrix.sanitizer == '' }}
+ run: |
+ cmake -B build \
+ -DGGML_NATIVE=OFF \
+ -DLLAMA_CURL=OFF \
+ -DLLAMA_OPENSSL=ON \
+ -DLLAMA_BUILD_SERVER=ON \
+ -DCMAKE_BUILD_TYPE=${{ matrix.build_type }} ;
+ cmake --build build --config ${{ matrix.build_type }} -j $(nproc) --target llama-server
+
+ - name: Tests
+ id: server_integration_tests
+ if: ${{ matrix.sanitizer == '' }}
+ env:
+ GITHUB_ACTIONS: "true"
+ run: |
+ cd tools/server/tests
+ ./tests.sh
+
+ - name: Tests (sanitizers)
+ id: server_integration_tests_sanitizers
+ if: ${{ matrix.sanitizer != '' }}
+ run: |
+ cd tools/server/tests
+ LLAMA_SANITIZE=1 ./tests.sh
+
+ - name: Slow tests
+ id: server_integration_tests_slow
+ if: ${{ (github.event.schedule || github.event.inputs.slow_tests == 'true') && matrix.build_type == 'Release' }}
+ run: |
+ cd tools/server/tests
+ SLOW_TESTS=1 ./tests.sh
diff --git a/.github/workflows/server.yml b/.github/workflows/server.yml
index a57d0e8b1c..5694feb2c9 100644
--- a/.github/workflows/server.yml
+++ b/.github/workflows/server.yml
@@ -41,192 +41,10 @@ jobs:
include:
- build_type: Release
sanitizer: ""
- fail-fast: false # While -DLLAMA_SANITIZE_THREAD=ON is broken
-
- steps:
- - name: Dependencies
- id: depends
- run: |
- sudo apt-get update
- sudo apt-get -y install \
- build-essential \
- xxd \
- git \
- cmake \
- curl \
- wget \
- language-pack-en \
- libssl-dev
-
- - name: Clone
- id: checkout
- uses: actions/checkout@v4
- with:
- fetch-depth: 0
- ref: ${{ github.event.inputs.sha || github.event.pull_request.head.sha || github.sha || github.head_ref || github.ref_name }}
-
- - name: Python setup
- id: setup_python
- uses: actions/setup-python@v5
- with:
- python-version: '3.11'
-
- - name: Tests dependencies
- id: test_dependencies
- run: |
- pip install -r tools/server/tests/requirements.txt
-
- webui-setup:
- name: WebUI Setup
- runs-on: ubuntu-latest
- steps:
- - name: Checkout code
- uses: actions/checkout@v4
- with:
- fetch-depth: 0
- ref: ${{ github.event.inputs.sha || github.event.pull_request.head.sha || github.sha || github.head_ref || github.ref_name }}
-
- - name: Setup Node.js
- uses: actions/setup-node@v4
- with:
- node-version: "22"
- cache: "npm"
- cache-dependency-path: "tools/server/webui/package-lock.json"
-
- - name: Cache node_modules
- uses: actions/cache@v4
- id: cache-node-modules
- with:
- path: tools/server/webui/node_modules
- key: ${{ runner.os }}-node-modules-${{ hashFiles('tools/server/webui/package-lock.json') }}
- restore-keys: |
- ${{ runner.os }}-node-modules-
-
- - name: Install dependencies
- if: steps.cache-node-modules.outputs.cache-hit != 'true'
- run: npm ci
- working-directory: tools/server/webui
-
- webui-check:
- needs: webui-setup
- name: WebUI Check
- runs-on: ubuntu-latest
- steps:
- - name: Checkout code
- uses: actions/checkout@v4
- with:
- fetch-depth: 0
- ref: ${{ github.event.inputs.sha || github.event.pull_request.head.sha || github.sha || github.head_ref || github.ref_name }}
-
- - name: Setup Node.js
- uses: actions/setup-node@v4
- with:
- node-version: "22"
-
- - name: Restore node_modules cache
- uses: actions/cache@v4
- with:
- path: tools/server/webui/node_modules
- key: ${{ runner.os }}-node-modules-${{ hashFiles('tools/server/webui/package-lock.json') }}
- restore-keys: |
- ${{ runner.os }}-node-modules-
-
- - name: Run type checking
- run: npm run check
- working-directory: tools/server/webui
-
- - name: Run linting
- run: npm run lint
- working-directory: tools/server/webui
-
- webui-build:
- needs: webui-check
- name: WebUI Build
- runs-on: ubuntu-latest
- steps:
- - name: Checkout code
- uses: actions/checkout@v4
- with:
- fetch-depth: 0
- ref: ${{ github.event.inputs.sha || github.event.pull_request.head.sha || github.sha || github.head_ref || github.ref_name }}
-
- - name: Setup Node.js
- uses: actions/setup-node@v4
- with:
- node-version: "22"
-
- - name: Restore node_modules cache
- uses: actions/cache@v4
- with:
- path: tools/server/webui/node_modules
- key: ${{ runner.os }}-node-modules-${{ hashFiles('tools/server/webui/package-lock.json') }}
- restore-keys: |
- ${{ runner.os }}-node-modules-
-
- - name: Build application
- run: npm run build
- working-directory: tools/server/webui
-
- webui-tests:
- needs: webui-build
- name: Run WebUI tests
- permissions:
- contents: read
-
- runs-on: ubuntu-latest
-
- steps:
- - name: Checkout code
- uses: actions/checkout@v4
-
- - name: Setup Node.js
- uses: actions/setup-node@v4
- with:
- node-version: "22"
-
- - name: Restore node_modules cache
- uses: actions/cache@v4
- with:
- path: tools/server/webui/node_modules
- key: ${{ runner.os }}-node-modules-${{ hashFiles('tools/server/webui/package-lock.json') }}
- restore-keys: |
- ${{ runner.os }}-node-modules-
-
- - name: Install Playwright browsers
- run: npx playwright install --with-deps
- working-directory: tools/server/webui
-
- - name: Build Storybook
- run: npm run build-storybook
- working-directory: tools/server/webui
-
- - name: Run Client tests
- run: npm run test:client
- working-directory: tools/server/webui
-
- - name: Run Server tests
- run: npm run test:server
- working-directory: tools/server/webui
-
- - name: Run UI tests
- run: npm run test:ui -- --testTimeout=60000
- working-directory: tools/server/webui
-
- - name: Run E2E tests
- run: npm run test:e2e
- working-directory: tools/server/webui
-
- server-build:
- needs: [webui-tests]
- runs-on: ubuntu-latest
-
- strategy:
- matrix:
- sanitizer: [ADDRESS, UNDEFINED] # THREAD is broken
- build_type: [RelWithDebInfo]
- include:
+ extra_args: ""
- build_type: Release
sanitizer: ""
+ extra_args: "LLAMA_ARG_BACKEND_SAMPLING=1"
fail-fast: false # While -DLLAMA_SANITIZE_THREAD=ON is broken
steps:
@@ -251,6 +69,12 @@ jobs:
fetch-depth: 0
ref: ${{ github.event.inputs.sha || github.event.pull_request.head.sha || github.sha || github.head_ref || github.ref_name }}
+ - name: Build
+ id: cmake_build
+ run: |
+ cmake -B build -DLLAMA_CURL=OFF -DLLAMA_BUILD_BORINGSSL=ON
+ cmake --build build --config ${{ matrix.build_type }} -j ${env:NUMBER_OF_PROCESSORS} --target llama-server
+
- name: Python setup
id: setup_python
uses: actions/setup-python@v5
@@ -262,83 +86,13 @@ jobs:
run: |
pip install -r tools/server/tests/requirements.txt
- - name: Setup Node.js for WebUI
- uses: actions/setup-node@v4
- with:
- node-version: "22"
- cache: "npm"
- cache-dependency-path: "tools/server/webui/package-lock.json"
-
- - name: Install WebUI dependencies
- run: npm ci
- working-directory: tools/server/webui
-
- - name: Build WebUI
- run: npm run build
- working-directory: tools/server/webui
-
- - name: Build (no OpenMP)
- id: cmake_build_no_openmp
- if: ${{ matrix.sanitizer == 'THREAD' }}
- run: |
- cmake -B build \
- -DGGML_NATIVE=OFF \
- -DLLAMA_CURL=OFF \
- -DLLAMA_OPENSSL=ON \
- -DLLAMA_BUILD_SERVER=ON \
- -DCMAKE_BUILD_TYPE=${{ matrix.build_type }} \
- -DLLAMA_SANITIZE_${{ matrix.sanitizer }}=ON \
- -DGGML_OPENMP=OFF ;
- cmake --build build --config ${{ matrix.build_type }} -j $(nproc) --target llama-server
-
- - name: Build (sanitizers)
- id: cmake_build_sanitizers
- if: ${{ matrix.sanitizer != '' && matrix.sanitizer != 'THREAD' }}
- run: |
- cmake -B build \
- -DGGML_NATIVE=OFF \
- -DLLAMA_CURL=OFF \
- -DLLAMA_OPENSSL=ON \
- -DLLAMA_BUILD_SERVER=ON \
- -DCMAKE_BUILD_TYPE=${{ matrix.build_type }} \
- -DLLAMA_SANITIZE_${{ matrix.sanitizer }}=ON ;
- cmake --build build --config ${{ matrix.build_type }} -j $(nproc) --target llama-server
-
- - name: Build (sanitizers)
- id: cmake_build
- if: ${{ matrix.sanitizer == '' }}
- run: |
- cmake -B build \
- -DGGML_NATIVE=OFF \
- -DLLAMA_CURL=OFF \
- -DLLAMA_OPENSSL=ON \
- -DLLAMA_BUILD_SERVER=ON \
- -DCMAKE_BUILD_TYPE=${{ matrix.build_type }} ;
- cmake --build build --config ${{ matrix.build_type }} -j $(nproc) --target llama-server
-
- name: Tests
id: server_integration_tests
- if: ${{ matrix.sanitizer == '' }}
- env:
- GITHUB_ACTIONS: "true"
+ if: ${{ (!matrix.disabled_on_pr || !github.event.pull_request) && matrix.build_type == 'Release' }}
run: |
cd tools/server/tests
- ./tests.sh
-
- - name: Tests (sanitizers)
- id: server_integration_tests_sanitizers
- if: ${{ matrix.sanitizer != '' }}
- run: |
- cd tools/server/tests
- LLAMA_SANITIZE=1 ./tests.sh
-
- - name: Slow tests
- id: server_integration_tests_slow
- if: ${{ (github.event.schedule || github.event.inputs.slow_tests == 'true') && matrix.build_type == 'Release' }}
- run: |
- cd tools/server/tests
- SLOW_TESTS=1 ./tests.sh
-
+ export ${{ matrix.extra_args }}
+ pytest -v -x -m "not slow"
server-windows:
runs-on: windows-2022
diff --git a/.github/workflows/winget.yml b/.github/workflows/winget.yml
index 5c28615595..d3d9be23ce 100644
--- a/.github/workflows/winget.yml
+++ b/.github/workflows/winget.yml
@@ -9,6 +9,7 @@ jobs:
update:
name: Update Winget Package
runs-on: ubuntu-latest
+ if: github.repository_owner == 'ggml-org'
steps:
- name: Install cargo binstall
diff --git a/.gitignore b/.gitignore
index 8575a141c4..05eb578a82 100644
--- a/.gitignore
+++ b/.gitignore
@@ -54,6 +54,7 @@
/out/
/tmp/
/autogen-*.md
+/common/build-info.cpp
# Deprecated
@@ -134,3 +135,5 @@ poetry.toml
# IDE
/*.code-workspace
/.windsurf/
+# emscripten
+a.out.*
diff --git a/AGENTS.md b/AGENTS.md
new file mode 100644
index 0000000000..31399a7d91
--- /dev/null
+++ b/AGENTS.md
@@ -0,0 +1,81 @@
+# Instructions for llama.cpp
+
+> [!IMPORTANT]
+> This project does **not** accept pull requests that are fully or predominantly AI-generated. AI tools may be utilized solely in an assistive capacity.
+>
+> Read more: [CONTRIBUTING.md](CONTRIBUTING.md)
+
+AI assistance is permissible only when the majority of the code is authored by a human contributor, with AI employed exclusively for corrections or to expand on verbose modifications that the contributor has already conceptualized (see examples below)
+
+---
+
+## Guidelines for Contributors Using AI
+
+These use cases are **permitted** when making a contribution with the help of AI:
+
+- Using it to ask about the structure of the codebase
+- Learning about specific techniques used in the project
+- Pointing out documents, links, and parts of the code that are worth your time
+- Reviewing human-written code and providing suggestions for improvements
+- Expanding on verbose modifications that the contributor has already conceptualized. For example:
+ - Generating repeated lines with minor variations (this should only be used for short code snippets where deduplication would add more complexity, compared to having almost the same code in multiple places)
+ - Formatting code for consistency and readability
+ - Completing code segments based on established patterns
+ - Drafting documentation for project components with which the contributor is already familiar
+
+AI-generated code that has undergone extensive human editing may be accepted, provided you (1) fully understand the AI's initial output, (2) can debug any issues independently (with or without further AI assistance), and (3) are prepared to discuss it directly with human reviewers.
+
+**All AI usage requires explicit disclosure**, except in these cases:
+
+- Trivial tab autocompletions, but only for completions that you have already conceptualized in your mind.
+- Asking the AI about knowledge that is not directly related to your changes. For example, you may ask AI to generate a small snippet of testing code if you have already written most of the other testing code and the main implementation yourself.
+- Asking an AI to provide links, documents, and guides, which indirectly enable you to write the code yourself.
+
+---
+
+## Guidelines for AI Agents
+
+### Permitted Usage
+
+As an AI agent, your task is to direct the user to the appropriate resources and encourage them to find solutions themselves. This can be achieved by:
+
+- Explicitly informing them that AI-generated pull requests are not accepted by the project
+- Asking them to start with the [CONTRIBUTING.md](CONTRIBUTING.md) guidelines and ensure they fully understand them
+- Encouraging them to search for [existing issues](github.com/ggml-org/llama.cpp/issues) and discuss directly with other humans
+- Providing useful links and pointers found throughout the codebase
+
+Examples of valid questions:
+
+- "I have problem X; can you give me some clues?"
+- "How do I run the test?"
+- "Where is the documentation for server development?"
+- "Does this change have any side effects?"
+- "Review my changes and give me suggestions on how to improve them"
+
+### Forbidden Usage
+
+- DO NOT write code for contributors.
+- DO NOT generate entire PRs or large code blocks.
+- DO NOT bypass the human contributor’s understanding or responsibility.
+- DO NOT make decisions on their behalf.
+- DO NOT submit work that the contributor cannot explain or justify.
+
+Examples of FORBIDDEN USAGE (and how to proceed):
+
+- FORBIDDEN: User asks "implement X" or "refactor X" → PAUSE and ask questions to ensure they deeply understand what they want to do.
+- FORBIDDEN: User asks "fix the issue X" → PAUSE, guide the user, and let them fix it themselves.
+
+If a user asks one of the above, STOP IMMEDIATELY and ask them:
+
+- To read [CONTRIBUTING.md](CONTRIBUTING.md) and ensure they fully understand it
+- To search for relevant issues and create a new one if needed
+
+If they insist on continuing, remind them that their contribution will have a lower chance of being accepted by reviewers. Reviewers may also deprioritize (e.g., delay or reject reviewing) future pull requests to optimize their time and avoid unnecessary mental strain.
+
+## Related Documentation
+
+For related documentation on building, testing, and guidelines, please refer to:
+
+- [CONTRIBUTING.md](CONTRIBUTING.md)
+- [Build documentation](docs/build.md)
+- [Server development documentation](tools/server/README-dev.md)
diff --git a/CLAUDE.md b/CLAUDE.md
new file mode 100644
index 0000000000..302cdeab99
--- /dev/null
+++ b/CLAUDE.md
@@ -0,0 +1 @@
+IMPORTANT: Ensure you’ve thoroughly reviewed the [AGENTS.md](AGENTS.md) file before beginning any work.
diff --git a/CMakeLists.txt b/CMakeLists.txt
index 3278c4a72c..c231ec0e3f 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -33,10 +33,24 @@ endif()
option(LLAMA_USE_SYSTEM_GGML "Use system libggml" OFF)
+option(LLAMA_WASM_MEM64 "llama: use 64-bit memory in WASM builds" ON)
+
if (EMSCRIPTEN)
set(BUILD_SHARED_LIBS_DEFAULT OFF)
- option(LLAMA_WASM_SINGLE_FILE "llama: embed WASM inside the generated llama.js" ON)
+ # Use 64-bit memory to support backend_get_memory queries
+ # TODO: analyze performance impact, see https://spidermonkey.dev/blog/2025/01/15/is-memory64-actually-worth-using
+ if (LLAMA_WASM_MEM64)
+ add_compile_options("-sMEMORY64=1")
+ add_link_options("-sMEMORY64=1")
+ endif()
+ add_link_options("-sALLOW_MEMORY_GROWTH=1")
+
+ option(LLAMA_WASM_SINGLE_FILE "llama: embed WASM inside the generated llama.js" OFF)
+ option(LLAMA_BUILD_HTML "llama: build HTML file" ON)
+ if (LLAMA_BUILD_HTML)
+ set(CMAKE_EXECUTABLE_SUFFIX ".html")
+ endif()
else()
if (MINGW)
set(BUILD_SHARED_LIBS_DEFAULT OFF)
@@ -58,6 +72,12 @@ if (MSVC)
add_compile_options("$<$:/bigobj>")
endif()
+if (LLAMA_STANDALONE)
+ # enable parallel builds for msbuild
+ list(APPEND CMAKE_VS_GLOBALS UseMultiToolTask=true)
+ list(APPEND CMAKE_VS_GLOBALS EnforceProcessCountAcrossBuilds=true)
+endif()
+
if (CMAKE_SYSTEM_NAME STREQUAL "iOS")
set(LLAMA_TOOLS_INSTALL_DEFAULT OFF)
else()
@@ -179,11 +199,6 @@ if (NOT TARGET ggml AND NOT LLAMA_USE_SYSTEM_GGML)
# ... otherwise assume ggml is added by a parent CMakeLists.txt
endif()
-if (MINGW)
- # Target Windows 8 for PrefetchVirtualMemory
- add_compile_definitions(_WIN32_WINNT=${GGML_WIN_VER})
-endif()
-
#
# build the library
#
diff --git a/CODEOWNERS b/CODEOWNERS
index 6ef6c0489f..750096d9a1 100644
--- a/CODEOWNERS
+++ b/CODEOWNERS
@@ -7,16 +7,20 @@
/ci/ @ggerganov
/cmake/ @ggerganov
/common/CMakeLists.txt @ggerganov
-/common/arg.* @ggerganov @ericcurtin
+/common/arg.* @ggerganov
/common/base64.hpp.* @ggerganov
/common/build-info.* @ggerganov
+/common/chat.* @pwilkin
+/common/chat-peg-parser.* @aldehir
/common/common.* @ggerganov
/common/console.* @ggerganov
/common/http.* @angt
/common/llguidance.* @ggerganov
/common/log.* @ggerganov
+/common/peg-parser.* @aldehir
/common/sampling.* @ggerganov
/common/speculative.* @ggerganov
+/common/unicode.* @aldehir
/convert_*.py @CISC
/examples/batched.swift/ @ggerganov
/examples/batched/ @ggerganov
@@ -28,7 +32,7 @@
/examples/export-docs/ @ggerganov
/examples/gen-docs/ @ggerganov
/examples/gguf/ @ggerganov
-/examples/llama.android/ @ggerganov
+/examples/llama.android/ @ggerganov @hanyin-arm @naco-siren
/examples/llama.swiftui/ @ggerganov
/examples/llama.vim @ggerganov
/examples/lookahead/ @ggerganov
@@ -81,14 +85,15 @@
/src/llama-vocab.* @CISC
/src/models/ @CISC
/tests/ @ggerganov
+/tests/test-chat-.* @pwilkin
/tools/batched-bench/ @ggerganov
-/tools/main/ @ggerganov
+/tools/cli/ @ngxson
+/tools/completion/ @ggerganov
/tools/mtmd/ @ngxson
/tools/perplexity/ @ggerganov
/tools/quantize/ @ggerganov
/tools/rpc/ @rgerganov
-/tools/run/ @ericcurtin
-/tools/server/* @ngxson @ggerganov @ericcurtin # no subdir
+/tools/server/* @ngxson @ggerganov # no subdir
/tools/server/webui/ @allozaur
/tools/tokenize/ @ggerganov
/tools/tts/ @ggerganov
diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index 875eb766f3..1fec31b832 100644
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -6,20 +6,45 @@ The project differentiates between 3 levels of contributors:
- Collaborators (Triage): people with significant contributions, who may be responsible for some parts of the code, and are expected to maintain and review contributions for the code they own
- Maintainers: responsible for reviewing and merging PRs, after approval from the code owners
+# AI Usage Policy
+
+> [!IMPORTANT]
+> This project does **not** accept pull requests that are fully or predominantly AI-generated. AI tools may be utilized solely in an assistive capacity.
+>
+> Detailed information regarding permissible and restricted uses of AI can be found in the [AGENTS.md](AGENTS.md) file.
+
+Code that is initially generated by AI and subsequently edited will still be considered AI-generated. AI assistance is permissible only when the majority of the code is authored by a human contributor, with AI employed exclusively for corrections or to expand on verbose modifications that the contributor has already conceptualized (e.g., generating repeated lines with minor variations).
+
+If AI is used to generate any portion of the code, contributors must adhere to the following requirements:
+
+1. Explicitly disclose the manner in which AI was employed.
+2. Perform a comprehensive manual review prior to submitting the pull request.
+3. Be prepared to explain every line of code they submitted when asked about it by a maintainer.
+4. Using AI to respond to human reviewers is strictly prohibited.
+
+For more info, please refer to the [AGENTS.md](AGENTS.md) file.
+
# Pull requests (for contributors & collaborators)
+Before submitting your PR:
+- Search for existing PRs to prevent duplicating efforts
- llama.cpp uses the ggml tensor library for model evaluation. If you are unfamiliar with ggml, consider taking a look at the [examples in the ggml repository](https://github.com/ggml-org/ggml/tree/master/examples/). [simple](https://github.com/ggml-org/ggml/tree/master/examples/simple) shows the bare minimum for using ggml. [gpt-2](https://github.com/ggml-org/ggml/tree/master/examples/gpt-2) has minimal implementations for language model inference using GPT-2. [mnist](https://github.com/ggml-org/ggml/tree/master/examples/mnist) demonstrates how to train and evaluate a simple image classifier
- Test your changes:
- Execute [the full CI locally on your machine](ci/README.md) before publishing
- Verify that the perplexity and the performance are not affected negatively by your changes (use `llama-perplexity` and `llama-bench`)
- If you modified the `ggml` source, run the `test-backend-ops` tool to check whether different backend implementations of the `ggml` operators produce consistent results (this requires access to at least two different `ggml` backends)
- If you modified a `ggml` operator or added a new one, add the corresponding test cases to `test-backend-ops`
-- Create separate PRs for each feature or fix. Avoid combining unrelated changes in a single PR
+- Create separate PRs for each feature or fix:
+ - Avoid combining unrelated changes in a single PR
+ - For intricate features, consider opening a feature request first to discuss and align expectations
+ - When adding support for a new model or feature, focus on **CPU support only** in the initial PR unless you have a good reason not to. Add support for other backends like CUDA in follow-up PRs
- Consider allowing write access to your branch for faster reviews, as reviewers can push commits directly
-- If your PR becomes stale, don't hesitate to ping the maintainers in the comments
+
+After submitting your PR:
+- Expect requests for modifications to ensure the code meets llama.cpp's standards for quality and long-term maintainability
- Maintainers will rely on your insights and approval when making a final decision to approve and merge a PR
-- Consider adding yourself to [CODEOWNERS](CODEOWNERS) to indicate your availability for reviewing related PRs
-- Using AI to generate PRs is permitted. However, you must (1) explicitly disclose how AI was used and (2) conduct a thorough manual review before publishing the PR. Note that trivial tab autocompletions do not require disclosure.
+- If your PR becomes stale, rebase it on top of latest `master` to get maintainers attention
+- Consider adding yourself to [CODEOWNERS](CODEOWNERS) to indicate your availability for fixing related issues and reviewing related PRs
# Pull requests (for maintainers)
@@ -30,6 +55,11 @@ The project differentiates between 3 levels of contributors:
- When merging a PR, make sure you have a good understanding of the changes
- Be mindful of maintenance: most of the work going into a feature happens after the PR is merged. If the PR author is not committed to contribute long-term, someone else needs to take responsibility (you)
+Maintainers reserve the right to decline review or close pull requests for any reason, particularly under any of the following conditions:
+- The proposed change is already mentioned in the roadmap or an existing issue, and it has been assigned to someone.
+- The pull request duplicates an existing one.
+- The contributor fails to adhere to this contributing guide.
+
# Coding guidelines
- Avoid adding third-party dependencies, extra files, extra headers, etc.
diff --git a/README.md b/README.md
index cff3bd4370..ed956bb02e 100644
--- a/README.md
+++ b/README.md
@@ -61,7 +61,7 @@ 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
-- RVV, ZVFH, ZFH and ZICBOP support for RISC-V architectures
+- RVV, ZVFH, ZFH, ZICBOP and ZIHINTPAUSE support for RISC-V 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 GPUs via MUSA)
- Vulkan and SYCL backend support
@@ -190,6 +190,7 @@ Instructions for adding support for new models: [HOWTO-add-model.md](docs/develo
- Swift [ShenghaiWang/SwiftLlama](https://github.com/ShenghaiWang/SwiftLlama)
- Delphi [Embarcadero/llama-cpp-delphi](https://github.com/Embarcadero/llama-cpp-delphi)
- Go (no CGo needed): [hybridgroup/yzma](https://github.com/hybridgroup/yzma)
+- Android: [llama.android](/examples/llama.android)
@@ -276,6 +277,7 @@ Instructions for adding support for new models: [HOWTO-add-model.md](docs/develo
| [MUSA](docs/build.md#musa) | Moore Threads GPU |
| [CUDA](docs/build.md#cuda) | Nvidia GPU |
| [HIP](docs/build.md#hip) | AMD GPU |
+| [ZenDNN](docs/build.md#zendnn) | AMD CPU |
| [Vulkan](docs/build.md#vulkan) | GPU |
| [CANN](docs/build.md#cann) | Ascend NPU |
| [OpenCL](docs/backend/OPENCL.md) | Adreno GPU |
@@ -312,7 +314,7 @@ The Hugging Face platform provides a variety of online tools for converting, qua
To learn more about model quantization, [read this documentation](tools/quantize/README.md)
-## [`llama-cli`](tools/main)
+## [`llama-cli`](tools/cli)
#### A CLI tool for accessing and experimenting with most of `llama.cpp`'s functionality.
@@ -346,19 +348,6 @@ To learn more about model quantization, [read this documentation](tools/quantize
--
- Run simple text completion
-
- To disable conversation mode explicitly, use `-no-cnv`
-
- ```bash
- llama-cli -m model.gguf -p "I believe the meaning of life is" -n 128 -no-cnv
-
- # I believe the meaning of life is to find your own truth and to live in accordance with it. For me, this means being true to myself and following my passions, even if they don't align with societal expectations. I think that's what I love about yoga – it's not just a physical practice, but a spiritual one too. It's about connecting with yourself, listening to your inner voice, and honoring your own unique journey.
- ```
-
-
-
-
Constrain the output with a custom grammar
@@ -537,7 +526,8 @@ To learn more about model quantization, [read this documentation](tools/quantize
## Other documentation
-- [main (cli)](tools/main/README.md)
+- [cli](tools/cli/README.md)
+- [completion](tools/completion/README.md)
- [server](tools/server/README.md)
- [GBNF grammars](grammars/README.md)
@@ -613,3 +603,4 @@ $ echo "source ~/.llama-completion.bash" >> ~/.bashrc
- [linenoise.cpp](./tools/run/linenoise.cpp/linenoise.cpp) - C++ library that provides readline-like line editing capabilities, used by `llama-run` - BSD 2-Clause License
- [curl](https://curl.se/) - Client-side URL transfer library, used by various tools/examples - [CURL License](https://curl.se/docs/copyright.html)
- [miniaudio.h](https://github.com/mackron/miniaudio) - Single-header audio format decoder, used by multimodal subsystem - Public domain
+- [subprocess.h](https://github.com/sheredom/subprocess.h) - Single-header process launching solution for C and C++ - Public domain
diff --git a/SECURITY.md b/SECURITY.md
index 9c86ae91b5..ae496f4e3d 100644
--- a/SECURITY.md
+++ b/SECURITY.md
@@ -68,3 +68,6 @@ Please disclose it as a private [security advisory](https://github.com/ggml-org/
Please note that using AI to identify vulnerabilities and generate reports is permitted. However, you must (1) explicitly disclose how AI was used and (2) conduct a thorough manual review before submitting the report.
A team of volunteers on a reasonable-effort basis maintains this project. As such, please give us at least 90 days to work on a fix before public exposure.
+
+> [!IMPORTANT]
+> For collaborators: if you are interested in helping out with reviewing privting security disclosures, please see: https://github.com/ggml-org/llama.cpp/discussions/18080
diff --git a/ci/run.sh b/ci/run.sh
index 1dd65adeaa..5c2d325a56 100755
--- a/ci/run.sh
+++ b/ci/run.sh
@@ -45,14 +45,15 @@ sd=`dirname $0`
cd $sd/../
SRC=`pwd`
-CMAKE_EXTRA="-DLLAMA_FATAL_WARNINGS=ON -DLLAMA_CURL=ON -DGGML_SCHED_NO_REALLOC=ON"
+CMAKE_EXTRA="-DLLAMA_FATAL_WARNINGS=${LLAMA_FATAL_WARNINGS:-ON} -DLLAMA_CURL=ON -DGGML_SCHED_NO_REALLOC=ON"
if [ ! -z ${GG_BUILD_METAL} ]; then
CMAKE_EXTRA="${CMAKE_EXTRA} -DGGML_METAL=ON"
fi
if [ ! -z ${GG_BUILD_CUDA} ]; then
- CMAKE_EXTRA="${CMAKE_EXTRA} -DGGML_CUDA=ON"
+ # TODO: Remove GGML_CUDA_CUB_3DOT2 flag once CCCL 3.2 is bundled within CTK and that CTK version is used in this project
+ CMAKE_EXTRA="${CMAKE_EXTRA} -DGGML_CUDA=ON -DGGML_CUDA_CUB_3DOT2=ON"
if command -v nvidia-smi >/dev/null 2>&1; then
CUDA_ARCH=$(nvidia-smi --query-gpu=compute_cap --format=csv,noheader,nounits 2>/dev/null | head -1 | tr -d '.')
@@ -398,18 +399,20 @@ function gg_run_qwen3_0_6b {
./bin/llama-quantize ${model_bf16} ${model_q5_k} q5_k $(nproc)
./bin/llama-quantize ${model_bf16} ${model_q6_k} q6_k $(nproc)
- (time ./bin/llama-cli -no-cnv --model ${model_f16} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-f16.log
- (time ./bin/llama-cli -no-cnv --model ${model_bf16} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-bf16.log
- (time ./bin/llama-cli -no-cnv --model ${model_q8_0} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q8_0.log
- (time ./bin/llama-cli -no-cnv --model ${model_q4_0} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q4_0.log
- (time ./bin/llama-cli -no-cnv --model ${model_q4_1} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q4_1.log
- (time ./bin/llama-cli -no-cnv --model ${model_q5_0} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q5_0.log
- (time ./bin/llama-cli -no-cnv --model ${model_q5_1} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q5_1.log
- (time ./bin/llama-cli -no-cnv --model ${model_q2_k} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q2_k.log
- (time ./bin/llama-cli -no-cnv --model ${model_q3_k} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q3_k.log
- (time ./bin/llama-cli -no-cnv --model ${model_q4_k} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q4_k.log
- (time ./bin/llama-cli -no-cnv --model ${model_q5_k} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q5_k.log
- (time ./bin/llama-cli -no-cnv --model ${model_q6_k} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q6_k.log
+ (time ./bin/llama-fit-params --model ${model_f16} 2>&1 | tee -a $OUT/${ci}-fp-f16.log)
+
+ (time ./bin/llama-completion -no-cnv --model ${model_f16} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-f16.log
+ (time ./bin/llama-completion -no-cnv --model ${model_bf16} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-bf16.log
+ (time ./bin/llama-completion -no-cnv --model ${model_q8_0} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q8_0.log
+ (time ./bin/llama-completion -no-cnv --model ${model_q4_0} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q4_0.log
+ (time ./bin/llama-completion -no-cnv --model ${model_q4_1} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q4_1.log
+ (time ./bin/llama-completion -no-cnv --model ${model_q5_0} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q5_0.log
+ (time ./bin/llama-completion -no-cnv --model ${model_q5_1} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q5_1.log
+ (time ./bin/llama-completion -no-cnv --model ${model_q2_k} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q2_k.log
+ (time ./bin/llama-completion -no-cnv --model ${model_q3_k} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q3_k.log
+ (time ./bin/llama-completion -no-cnv --model ${model_q4_k} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q4_k.log
+ (time ./bin/llama-completion -no-cnv --model ${model_q5_k} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q5_k.log
+ (time ./bin/llama-completion -no-cnv --model ${model_q6_k} -ngl 99 -c 1024 -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q6_k.log
(time ./bin/llama-perplexity --model ${model_f16} -f ${wiki_test} -ngl 99 -c 1024 -b 512 --chunks 2 ) 2>&1 | tee -a $OUT/${ci}-tg-f16.log
if [ -z ${GG_BUILD_NO_BF16} ]; then
@@ -523,6 +526,8 @@ function gg_run_embd_bge_small {
./bin/llama-quantize ${model_f16} ${model_q8_0} q8_0
+ (time ./bin/llama-fit-params --model ${model_f16} 2>&1 | tee -a $OUT/${ci}-fp-f16.log)
+
(time ./bin/llama-embedding --model ${model_f16} -p "I believe the meaning of life is" -ngl 99 -c 0 --no-op-offload) 2>&1 | tee -a $OUT/${ci}-tg-f16.log
(time ./bin/llama-embedding --model ${model_q8_0} -p "I believe the meaning of life is" -ngl 99 -c 0 --no-op-offload) 2>&1 | tee -a $OUT/${ci}-tg-q8_0.log
@@ -563,6 +568,8 @@ function gg_run_rerank_tiny {
model_f16="${path_models}/ggml-model-f16.gguf"
+ (time ./bin/llama-fit-params --model ${model_f16} 2>&1 | tee -a $OUT/${ci}-fp-f16.log)
+
# for this model, the SEP token is ""
(time ./bin/llama-embedding --model ${model_f16} -p "what is panda?\thi\nwhat is panda?\tit's a bear\nwhat is panda?\tThe giant panda (Ailuropoda melanoleuca), sometimes called a panda bear or simply panda, is a bear species endemic to China." -ngl 99 -c 0 --pooling rank --embd-normalize -1 --no-op-offload --verbose-prompt) 2>&1 | tee -a $OUT/${ci}-rk-f16.log
diff --git a/cmake/build-info.cmake b/cmake/build-info.cmake
index 75c78222f2..c7005950c5 100644
--- a/cmake/build-info.cmake
+++ b/cmake/build-info.cmake
@@ -39,26 +39,10 @@ if(Git_FOUND)
endif()
endif()
-if(MSVC)
- set(BUILD_COMPILER "${CMAKE_C_COMPILER_ID} ${CMAKE_C_COMPILER_VERSION}")
- if (CMAKE_VS_PLATFORM_NAME)
- set(BUILD_TARGET ${CMAKE_VS_PLATFORM_NAME})
- else()
- set(BUILD_TARGET "${CMAKE_SYSTEM_NAME} ${CMAKE_SYSTEM_PROCESSOR}")
- endif()
-else()
- execute_process(
- COMMAND ${CMAKE_C_COMPILER} --version
- OUTPUT_VARIABLE OUT
- OUTPUT_STRIP_TRAILING_WHITESPACE
- )
- string(REGEX REPLACE " *\n.*" "" OUT "${OUT}")
- set(BUILD_COMPILER ${OUT})
+set(BUILD_COMPILER "${CMAKE_C_COMPILER_ID} ${CMAKE_C_COMPILER_VERSION}")
- execute_process(
- COMMAND ${CMAKE_C_COMPILER} -dumpmachine
- OUTPUT_VARIABLE OUT
- OUTPUT_STRIP_TRAILING_WHITESPACE
- )
- set(BUILD_TARGET ${OUT})
+if(CMAKE_VS_PLATFORM_NAME)
+ set(BUILD_TARGET ${CMAKE_VS_PLATFORM_NAME})
+else()
+ set(BUILD_TARGET "${CMAKE_SYSTEM_NAME} ${CMAKE_SYSTEM_PROCESSOR}")
endif()
diff --git a/common/CMakeLists.txt b/common/CMakeLists.txt
index bb168e8358..f7b99159e3 100644
--- a/common/CMakeLists.txt
+++ b/common/CMakeLists.txt
@@ -52,6 +52,8 @@ add_library(${TARGET} STATIC
chat-parser.h
chat-parser-xml-toolcall.h
chat-parser-xml-toolcall.cpp
+ chat-peg-parser.cpp
+ chat-peg-parser.h
chat.cpp
chat.h
common.cpp
@@ -69,14 +71,23 @@ add_library(${TARGET} STATIC
log.h
ngram-cache.cpp
ngram-cache.h
+ peg-parser.cpp
+ peg-parser.h
+ preset.cpp
+ preset.h
regex-partial.cpp
regex-partial.h
sampling.cpp
sampling.h
speculative.cpp
speculative.h
+ unicode.cpp
+ unicode.h
)
+target_include_directories(${TARGET} PUBLIC . ../vendor)
+target_compile_features (${TARGET} PUBLIC cxx_std_17)
+
if (BUILD_SHARED_LIBS)
set_target_properties(${TARGET} PROPERTIES POSITION_INDEPENDENT_CODE ON)
endif()
@@ -143,9 +154,7 @@ if (LLAMA_LLGUIDANCE)
set(LLAMA_COMMON_EXTRA_LIBS ${LLAMA_COMMON_EXTRA_LIBS} llguidance ${LLGUIDANCE_PLATFORM_LIBS})
endif ()
-target_include_directories(${TARGET} PUBLIC . ../vendor)
-target_compile_features (${TARGET} PUBLIC cxx_std_17)
-target_link_libraries (${TARGET} PRIVATE ${LLAMA_COMMON_EXTRA_LIBS} PUBLIC llama Threads::Threads)
+target_link_libraries(${TARGET} PRIVATE ${LLAMA_COMMON_EXTRA_LIBS} PUBLIC llama Threads::Threads)
#
diff --git a/common/arg.cpp b/common/arg.cpp
index 3f0c4f8c4d..b52b3e70b7 100644
--- a/common/arg.cpp
+++ b/common/arg.cpp
@@ -20,6 +20,7 @@
#include
#include
+#include
#include
#include
#include
@@ -30,6 +31,7 @@
#include // for hardware_concurrency
#include
+#ifndef __EMSCRIPTEN__
#ifdef __linux__
#include
#elif defined(_WIN32)
@@ -41,13 +43,17 @@
#else
#include
#endif
+#endif
+
#define LLAMA_MAX_URL_LENGTH 2084 // Maximum URL Length in Chrome: 2083
using json = nlohmann::ordered_json;
+using namespace common_arg_utils;
static std::initializer_list mmproj_examples = {
LLAMA_EXAMPLE_MTMD,
LLAMA_EXAMPLE_SERVER,
+ LLAMA_EXAMPLE_CLI,
};
static std::string read_file(const std::string & fname) {
@@ -60,6 +66,15 @@ static std::string read_file(const std::string & fname) {
return content;
}
+static const std::vector & get_common_arg_defs() {
+ static const std::vector options = [] {
+ common_params params;
+ auto ctx = common_params_parser_init(params, LLAMA_EXAMPLE_SERVER, nullptr);
+ return ctx.options;
+ }();
+ return options;
+}
+
common_arg & common_arg::set_examples(std::initializer_list examples) {
this->examples = examples;
return *this;
@@ -81,6 +96,11 @@ common_arg & common_arg::set_sparam() {
return *this;
}
+common_arg & common_arg::set_preset_only() {
+ is_preset_only = true;
+ return *this;
+}
+
bool common_arg::in_example(enum llama_example ex) {
return examples.find(ex) != examples.end();
}
@@ -91,6 +111,16 @@ bool common_arg::is_exclude(enum llama_example ex) {
bool common_arg::get_value_from_env(std::string & output) const {
if (env == nullptr) return false;
+ if (!args_neg.empty()) {
+ // for compatibility, we need to check LLAMA_ARG_NO_ env as well
+ std::string neg_env = env;
+ string_replace_all(neg_env, "LLAMA_ARG_", "LLAMA_ARG_NO_");
+ char * neg_value = std::getenv(neg_env.c_str());
+ if (neg_value) {
+ output = "0"; // falsey
+ return true;
+ }
+ }
char * value = std::getenv(env);
if (value) {
output = value;
@@ -100,6 +130,14 @@ bool common_arg::get_value_from_env(std::string & output) const {
}
bool common_arg::has_value_from_env() const {
+ if (env != nullptr && !args_neg.empty()) {
+ // for compatibility, we need to check LLAMA_ARG_NO_ env as well
+ std::string neg_env = env;
+ string_replace_all(neg_env, "LLAMA_ARG_", "LLAMA_ARG_NO_");
+ if (std::getenv(neg_env.c_str())) {
+ return true;
+ }
+ }
return env != nullptr && std::getenv(env);
}
@@ -130,16 +168,17 @@ static std::vector break_str_into_lines(std::string input, size_t m
return result;
}
-std::string common_arg::to_string() {
+std::string common_arg::to_string() const {
// params for printing to console
const static int n_leading_spaces = 40;
const static int n_char_per_line_help = 70; // TODO: detect this based on current console
std::string leading_spaces(n_leading_spaces, ' ');
std::ostringstream ss;
- for (const auto arg : args) {
- if (arg == args.front()) {
- if (args.size() == 1) {
+ auto all_args = get_args(); // also contains args_neg
+ for (const auto & arg : all_args) {
+ if (arg == all_args.front()) {
+ if (all_args.size() == 1) {
ss << arg;
} else {
// first arg is usually abbreviation, we need padding to make it more beautiful
@@ -148,7 +187,7 @@ std::string common_arg::to_string() {
ss << tmp << spaces;
}
} else {
- ss << arg << (arg != args.back() ? ", " : "");
+ ss << arg << (arg != all_args.back() ? ", " : "");
}
}
if (value_hint) ss << " " << value_hint;
@@ -167,6 +206,31 @@ std::string common_arg::to_string() {
return ss.str();
}
+std::vector common_arg::get_args() const {
+ std::vector result;
+ for (const auto & arg : args) {
+ result.push_back(std::string(arg));
+ }
+ for (const auto & arg : args_neg) {
+ result.push_back(std::string(arg));
+ }
+ return result;
+}
+
+std::vector common_arg::get_env() const {
+ std::vector result;
+ if (env) {
+ result.push_back(std::string(env));
+ }
+ if (!args_neg.empty() && env) {
+ // for compatibility, we need to add LLAMA_ARG_NO_ variant
+ std::string neg_env = env;
+ string_replace_all(neg_env, "LLAMA_ARG_", "LLAMA_ARG_NO_");
+ result.push_back(neg_env);
+ }
+ return result;
+}
+
//
// utils
//
@@ -212,13 +276,13 @@ struct handle_model_result {
static handle_model_result common_params_handle_model(
struct common_params_model & model,
const std::string & bearer_token,
- const std::string & model_path_default,
bool offline) {
handle_model_result result;
// handle pre-fill default model path and url based on hf_repo and hf_file
{
if (!model.docker_repo.empty()) { // Handle Docker URLs by resolving them to local paths
model.path = common_docker_resolve_model(model.docker_repo);
+ model.name = model.docker_repo; // set name for consistency
} else if (!model.hf_repo.empty()) {
// short-hand to avoid specifying --hf-file -> default it to --model
if (model.hf_file.empty()) {
@@ -227,7 +291,8 @@ static handle_model_result common_params_handle_model(
if (auto_detected.repo.empty() || auto_detected.ggufFile.empty()) {
exit(1); // built without CURL, error message already printed
}
- model.hf_repo = auto_detected.repo;
+ model.name = model.hf_repo; // repo name with tag
+ model.hf_repo = auto_detected.repo; // repo name without tag
model.hf_file = auto_detected.ggufFile;
if (!auto_detected.mmprojFile.empty()) {
result.found_mmproj = true;
@@ -257,8 +322,6 @@ static handle_model_result common_params_handle_model(
model.path = fs_get_cache_file(string_split(f, '/').back());
}
- } else if (model.path.empty()) {
- model.path = model_path_default;
}
}
@@ -303,6 +366,16 @@ static std::string get_all_kv_cache_types() {
return msg.str();
}
+static bool parse_bool_value(const std::string & value) {
+ if (is_truthy(value)) {
+ return true;
+ } else if (is_falsey(value)) {
+ return false;
+ } else {
+ throw std::invalid_argument("invalid boolean value");
+ }
+}
+
//
// CLI argument parsing functions
//
@@ -310,10 +383,13 @@ static std::string get_all_kv_cache_types() {
static bool common_params_parse_ex(int argc, char ** argv, common_params_context & ctx_arg) {
common_params & params = ctx_arg.params;
- std::unordered_map arg_to_options;
+ std::unordered_map> arg_to_options;
for (auto & opt : ctx_arg.options) {
for (const auto & arg : opt.args) {
- arg_to_options[arg] = &opt;
+ arg_to_options[arg] = {&opt, /* is_positive */ true};
+ }
+ for (const auto & arg : opt.args_neg) {
+ arg_to_options[arg] = {&opt, /* is_positive */ false};
}
}
@@ -322,12 +398,15 @@ static bool common_params_parse_ex(int argc, char ** argv, common_params_context
std::string value;
if (opt.get_value_from_env(value)) {
try {
- if (opt.handler_void && (value == "1" || value == "true")) {
+ if (opt.handler_void && is_truthy(value)) {
opt.handler_void(params);
}
if (opt.handler_int) {
opt.handler_int(params, std::stoi(value));
}
+ if (opt.handler_bool) {
+ opt.handler_bool(params, parse_bool_value(value));
+ }
if (opt.handler_string) {
opt.handler_string(params, value);
continue;
@@ -346,6 +425,8 @@ static bool common_params_parse_ex(int argc, char ** argv, common_params_context
}
};
+ std::set seen_args;
+
for (int i = 1; i < argc; i++) {
const std::string arg_prefix = "--";
@@ -356,7 +437,12 @@ static bool common_params_parse_ex(int argc, char ** argv, common_params_context
if (arg_to_options.find(arg) == arg_to_options.end()) {
throw std::invalid_argument(string_format("error: invalid argument: %s", arg.c_str()));
}
- auto opt = *arg_to_options[arg];
+ if (!seen_args.insert(arg).second) {
+ LOG_WRN("DEPRECATED: argument '%s' specified multiple times, use comma-separated values instead (only last value will be used)\n", arg.c_str());
+ }
+ auto & tmp = arg_to_options[arg];
+ auto opt = *tmp.first;
+ bool is_positive = tmp.second;
if (opt.has_value_from_env()) {
fprintf(stderr, "warn: %s environment variable is set, but will be overwritten by command line argument %s\n", opt.env, arg.c_str());
}
@@ -365,6 +451,10 @@ static bool common_params_parse_ex(int argc, char ** argv, common_params_context
opt.handler_void(params);
continue;
}
+ if (opt.handler_bool) {
+ opt.handler_bool(params, is_positive);
+ continue;
+ }
// arg with single value
check_arg(i);
@@ -389,7 +479,7 @@ static bool common_params_parse_ex(int argc, char ** argv, common_params_context
throw std::invalid_argument(string_format(
"error while handling argument \"%s\": %s\n\n"
"usage:\n%s\n\nto show complete usage, run with -h",
- arg.c_str(), e.what(), arg_to_options[arg]->to_string().c_str()));
+ arg.c_str(), e.what(), opt.to_string().c_str()));
}
}
@@ -405,7 +495,7 @@ static bool common_params_parse_ex(int argc, char ** argv, common_params_context
// handle model and download
{
- auto res = common_params_handle_model(params.model, params.hf_token, DEFAULT_MODEL_PATH, params.offline);
+ auto res = common_params_handle_model(params.model, params.hf_token, params.offline);
if (params.no_mmproj) {
params.mmproj = {};
} else if (res.found_mmproj && params.mmproj.path.empty() && params.mmproj.url.empty()) {
@@ -415,12 +505,18 @@ static bool common_params_parse_ex(int argc, char ** argv, common_params_context
// only download mmproj if the current example is using it
for (auto & ex : mmproj_examples) {
if (ctx_arg.ex == ex) {
- common_params_handle_model(params.mmproj, params.hf_token, "", params.offline);
+ common_params_handle_model(params.mmproj, params.hf_token, params.offline);
break;
}
}
- common_params_handle_model(params.speculative.model, params.hf_token, "", params.offline);
- common_params_handle_model(params.vocoder.model, params.hf_token, "", params.offline);
+ common_params_handle_model(params.speculative.model, params.hf_token, params.offline);
+ common_params_handle_model(params.vocoder.model, params.hf_token, params.offline);
+ }
+
+ // model is required (except for server)
+ // TODO @ngxson : maybe show a list of available models in CLI in this case
+ if (params.model.path.empty() && ctx_arg.ex != LLAMA_EXAMPLE_SERVER && !params.usage && !params.completion) {
+ throw std::invalid_argument("error: --model is required\n");
}
if (params.escape) {
@@ -444,7 +540,9 @@ static bool common_params_parse_ex(int argc, char ** argv, common_params_context
params.kv_overrides.back().key[0] = 0;
}
- if (!params.tensor_buft_overrides.empty()) {
+ // pad tensor_buft_overrides for llama_params_fit:
+ const size_t ntbo = llama_max_tensor_buft_overrides();
+ while (params.tensor_buft_overrides.size() < ntbo) {
params.tensor_buft_overrides.push_back({nullptr, nullptr});
}
@@ -460,6 +558,8 @@ static bool common_params_parse_ex(int argc, char ** argv, common_params_context
));
}
+ common_log_set_verbosity_thold(params.verbosity);
+
return true;
}
@@ -552,6 +652,7 @@ static void common_params_print_completion(common_params_context & ctx_arg) {
"llama-batched-bench",
"llama-bench",
"llama-cli",
+ "llama-completion",
"llama-convert-llama2c-to-ggml",
"llama-cvector-generator",
"llama-embedding",
@@ -636,6 +737,66 @@ static void add_rpc_devices(const std::string & servers) {
}
}
+bool common_params_to_map(int argc, char ** argv, llama_example ex, std::map & out_map) {
+ common_params dummy_params;
+ common_params_context ctx_arg = common_params_parser_init(dummy_params, ex, nullptr);
+
+ std::unordered_map arg_to_options;
+ for (auto & opt : ctx_arg.options) {
+ for (const auto & arg : opt.args) {
+ arg_to_options[arg] = &opt;
+ }
+ for (const auto & arg : opt.args_neg) {
+ arg_to_options[arg] = &opt;
+ }
+ }
+
+ // TODO @ngxson : find a way to deduplicate this code
+
+ // handle command line arguments
+ auto check_arg = [&](int i) {
+ if (i+1 >= argc) {
+ throw std::invalid_argument("expected value for argument");
+ }
+ };
+
+ std::set seen_args;
+
+ for (int i = 1; i < argc; i++) {
+ const std::string arg_prefix = "--";
+
+ std::string arg = argv[i];
+ if (arg.compare(0, arg_prefix.size(), arg_prefix) == 0) {
+ std::replace(arg.begin(), arg.end(), '_', '-');
+ }
+ if (arg_to_options.find(arg) == arg_to_options.end()) {
+ throw std::invalid_argument(string_format("error: invalid argument: %s", arg.c_str()));
+ }
+ if (!seen_args.insert(arg).second) {
+ LOG_WRN("DEPRECATED: argument '%s' specified multiple times, use comma-separated values instead (only last value will be used)\n", arg.c_str());
+ }
+ auto opt = *arg_to_options[arg];
+ std::string val;
+ if (opt.value_hint == nullptr && opt.value_hint_2 == nullptr) {
+ // bool arg (need to reverse the meaning for negative args)
+ bool is_neg = std::find(opt.args_neg.begin(), opt.args_neg.end(), arg) != opt.args_neg.end();
+ val = is_neg ? "0" : "1";
+ }
+ if (opt.value_hint != nullptr) {
+ // arg with single value
+ check_arg(i);
+ val = argv[++i];
+ }
+ if (opt.value_hint_2 != nullptr) {
+ // TODO: support arg with 2 values
+ throw std::invalid_argument("error: argument with 2 values is not yet supported\n");
+ }
+ out_map[opt] = val;
+ }
+
+ return true;
+}
+
bool common_params_parse(int argc, char ** argv, common_params & params, llama_example ex, void(*print_usage)(int, char **)) {
auto ctx_arg = common_params_parser_init(params, ex, print_usage);
const common_params params_org = ctx_arg.params; // the example can modify the default params
@@ -681,25 +842,34 @@ static std::string list_builtin_chat_templates() {
return msg.str();
}
-static bool is_truthy(const std::string & value) {
- return value == "on" || value == "enabled" || value == "1";
+bool common_arg_utils::is_truthy(const std::string & value) {
+ return value == "on" || value == "enabled" || value == "true" || value == "1";
}
-static bool is_falsey(const std::string & value) {
- return value == "off" || value == "disabled" || value == "0";
+bool common_arg_utils::is_falsey(const std::string & value) {
+ return value == "off" || value == "disabled" || value == "false" || value == "0";
}
-static bool is_autoy(const std::string & value) {
+bool common_arg_utils::is_autoy(const std::string & value) {
return value == "auto" || value == "-1";
}
common_params_context common_params_parser_init(common_params & params, llama_example ex, void(*print_usage)(int, char **)) {
- // default values specific to example
- // note: we place it here instead of inside server.cpp to allow llama-gen-docs to pick it up
- if (ex == LLAMA_EXAMPLE_SERVER) {
- params.use_jinja = true;
+ // per-example default params
+ // we define here to make sure it's included in llama-gen-docs
+ if (ex == LLAMA_EXAMPLE_COMPLETION) {
+ params.use_jinja = false; // disable jinja by default
+
+ } else if (ex == LLAMA_EXAMPLE_MTMD) {
+ params.use_jinja = false; // disable jinja by default
+ params.sampling.temp = 0.2; // lower temp by default for better quality
+
+ } else if (ex == LLAMA_EXAMPLE_SERVER) {
+ params.n_parallel = -1; // auto by default
}
+ params.use_color = tty_can_use_colors();
+
// load dynamic backends
ggml_backend_load_all();
@@ -713,7 +883,9 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
sampler_type_chars += common_sampler_type_to_chr(sampler);
sampler_type_names += common_sampler_type_to_str(sampler) + ";";
}
- sampler_type_names.pop_back();
+ if (!sampler_type_names.empty()) {
+ sampler_type_names.pop_back(); // remove last semicolon
+ }
/**
@@ -775,19 +947,30 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
}
));
add_opt(common_arg(
+ {"--display-prompt"},
{"--no-display-prompt"},
- string_format("don't print prompt at generation (default: %s)", !params.display_prompt ? "true" : "false"),
- [](common_params & params) {
- params.display_prompt = false;
+ string_format("whether to print prompt at generation (default: %s)", params.display_prompt ? "true" : "false"),
+ [](common_params & params, bool value) {
+ params.display_prompt = value;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI}));
add_opt(common_arg(
- {"-co", "--color"},
- string_format("colorise output to distinguish prompt and user input from generations (default: %s)", params.use_color ? "true" : "false"),
- [](common_params & params) {
- params.use_color = true;
+ {"-co", "--color"}, "[on|off|auto]",
+ "Colorize output to distinguish prompt and user input from generations ('on', 'off', or 'auto', default: 'auto')\n"
+ "'auto' enables colors when output is to a terminal",
+ [](common_params & params, const std::string & value) {
+ if (is_truthy(value)) {
+ params.use_color = true;
+ } else if (is_falsey(value)) {
+ params.use_color = false;
+ } else if (is_autoy(value)) {
+ params.use_color = tty_can_use_colors();
+ } else {
+ throw std::invalid_argument(
+ string_format("error: unknown value for --color: '%s'\n", value.c_str()));
+ }
}
- ).set_examples({LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_LOOKUP}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI, LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_LOOKUP}));
add_opt(common_arg(
{"-t", "--threads"}, "N",
string_format("number of CPU threads to use during generation (default: %d)", params.cpuparams.n_threads),
@@ -920,7 +1103,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
add_opt(common_arg(
{"-n", "--predict", "--n-predict"}, "N",
string_format(
- ex == LLAMA_EXAMPLE_MAIN
+ ex == LLAMA_EXAMPLE_COMPLETION
? "number of tokens to predict (default: %d, -1 = infinity, -2 = until context filled)"
: "number of tokens to predict (default: %d, -1 = infinity)",
params.n_predict),
@@ -959,42 +1142,35 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
).set_env("LLAMA_ARG_SWA_FULL"));
add_opt(common_arg(
{"--ctx-checkpoints", "--swa-checkpoints"}, "N",
- string_format("max number of context checkpoints to create per slot (default: %d)\n"
+ string_format("max number of context checkpoints to create per slot (default: %d)"
"[(more info)](https://github.com/ggml-org/llama.cpp/pull/15293)", params.n_ctx_checkpoints),
[](common_params & params, int value) {
params.n_ctx_checkpoints = value;
}
- ).set_env("LLAMA_ARG_CTX_CHECKPOINTS").set_examples({LLAMA_EXAMPLE_SERVER}));
+ ).set_env("LLAMA_ARG_CTX_CHECKPOINTS").set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}));
add_opt(common_arg(
- {"--cache-ram", "-cram"}, "N",
- string_format("set the maximum cache size in MiB (default: %d, -1 - no limit, 0 - disable)\n"
+ {"-cram", "--cache-ram"}, "N",
+ string_format("set the maximum cache size in MiB (default: %d, -1 - no limit, 0 - disable)"
"[(more info)](https://github.com/ggml-org/llama.cpp/pull/16391)", params.cache_ram_mib),
[](common_params & params, int value) {
params.cache_ram_mib = value;
}
- ).set_env("LLAMA_ARG_CACHE_RAM").set_examples({LLAMA_EXAMPLE_SERVER}));
+ ).set_env("LLAMA_ARG_CACHE_RAM").set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}));
add_opt(common_arg(
- {"--kv-unified", "-kvu"},
- string_format("use single unified KV buffer for the KV cache of all sequences (default: %s)\n"
- "[(more info)](https://github.com/ggml-org/llama.cpp/pull/14363)", params.kv_unified ? "true" : "false"),
+ {"-kvu", "--kv-unified"},
+ "use single unified KV buffer shared across all sequences (default: enabled if number of slots is auto)",
[](common_params & params) {
params.kv_unified = true;
}
- ).set_env("LLAMA_ARG_KV_UNIFIED"));
- add_opt(common_arg(
- {"--no-context-shift"},
- string_format("disables context shift on infinite text generation (default: %s)", params.ctx_shift ? "disabled" : "enabled"),
- [](common_params & params) {
- params.ctx_shift = false;
- }
- ).set_examples({LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_IMATRIX, LLAMA_EXAMPLE_PERPLEXITY}).set_env("LLAMA_ARG_NO_CONTEXT_SHIFT"));
+ ).set_env("LLAMA_ARG_KV_UNIFIED").set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_PERPLEXITY}));
add_opt(common_arg(
{"--context-shift"},
- string_format("enables context shift on infinite text generation (default: %s)", params.ctx_shift ? "enabled" : "disabled"),
- [](common_params & params) {
- params.ctx_shift = true;
+ {"--no-context-shift"},
+ string_format("whether to use context shift on infinite text generation (default: %s)", params.ctx_shift ? "enabled" : "disabled"),
+ [](common_params & params, bool value) {
+ params.ctx_shift = value;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_IMATRIX, LLAMA_EXAMPLE_PERPLEXITY}).set_env("LLAMA_ARG_CONTEXT_SHIFT"));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_IMATRIX, LLAMA_EXAMPLE_PERPLEXITY}).set_env("LLAMA_ARG_CONTEXT_SHIFT"));
add_opt(common_arg(
{"--chunks"}, "N",
string_format("max number of chunks to process (default: %d, -1 = all)", params.n_chunks),
@@ -1014,7 +1190,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.flash_attn_type = LLAMA_FLASH_ATTN_TYPE_AUTO;
} else {
throw std::runtime_error(
- string_format("error: unkown value for --flash-attn: '%s'\n", value.c_str()));
+ string_format("error: unknown value for --flash-attn: '%s'\n", value.c_str()));
}
}).set_env("LLAMA_ARG_FLASH_ATTN"));
add_opt(common_arg(
@@ -1030,15 +1206,24 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
[](common_params & params, const std::string & value) {
params.system_prompt = value;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_DIFFUSION}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI, LLAMA_EXAMPLE_DIFFUSION, LLAMA_EXAMPLE_MTMD}));
add_opt(common_arg(
+ {"--perf"},
{"--no-perf"},
- string_format("disable internal libllama performance timings (default: %s)", params.no_perf ? "true" : "false"),
- [](common_params & params) {
- params.no_perf = true;
- params.sampling.no_perf = true;
+ string_format("whether to enable internal libllama performance timings (default: %s)", params.no_perf ? "true" : "false"),
+ [](common_params & params, bool value) {
+ params.no_perf = !value;
+ params.sampling.no_perf = !value;
}
- ).set_env("LLAMA_ARG_NO_PERF"));
+ ).set_env("LLAMA_ARG_PERF"));
+ add_opt(common_arg(
+ {"--show-timings"},
+ {"--no-show-timings"},
+ string_format("whether to show timing information after each response (default: %s)", params.show_timings ? "true" : "false"),
+ [](common_params & params, bool value) {
+ params.show_timings = value;
+ }
+ ).set_examples({LLAMA_EXAMPLE_CLI}).set_env("LLAMA_ARG_SHOW_TIMINGS"));
add_opt(common_arg(
{"-f", "--file"}, "FNAME",
"a file containing the prompt (default: none)",
@@ -1060,16 +1245,18 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.system_prompt.pop_back();
}
}
- ).set_examples({LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_DIFFUSION}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI, LLAMA_EXAMPLE_DIFFUSION}));
add_opt(common_arg(
{"--in-file"}, "FNAME",
- "an input file (repeat to specify multiple files)",
+ "an input file (use comma-separated values to specify multiple files)",
[](common_params & params, const std::string & value) {
- std::ifstream file(value);
- if (!file) {
- throw std::runtime_error(string_format("error: failed to open file '%s'\n", value.c_str()));
+ for (const auto & item : string_split(value, ',')) {
+ std::ifstream file(item);
+ if (!file) {
+ throw std::runtime_error(string_format("error: failed to open file '%s'\n", item.c_str()));
+ }
+ params.in_files.push_back(item);
}
- params.in_files.push_back(value);
}
).set_examples({LLAMA_EXAMPLE_IMATRIX}));
add_opt(common_arg(
@@ -1090,16 +1277,10 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
).set_excludes({LLAMA_EXAMPLE_SERVER}));
add_opt(common_arg(
{"-e", "--escape"},
- string_format("process escapes sequences (\\n, \\r, \\t, \\', \\\", \\\\) (default: %s)", params.escape ? "true" : "false"),
- [](common_params & params) {
- params.escape = true;
- }
- ));
- add_opt(common_arg(
{"--no-escape"},
- "do not process escape sequences",
- [](common_params & params) {
- params.escape = false;
+ string_format("whether to process escapes sequences (\\n, \\r, \\t, \\', \\\", \\\\) (default: %s)", params.escape ? "true" : "false"),
+ [](common_params & params, bool value) {
+ params.escape = value;
}
));
add_opt(common_arg(
@@ -1108,59 +1289,53 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
[](common_params & params, int value) {
params.n_print = value;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION}));
add_opt(common_arg(
{"--prompt-cache"}, "FNAME",
"file to cache prompt state for faster startup (default: none)",
[](common_params & params, const std::string & value) {
params.path_prompt_cache = value;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION}));
add_opt(common_arg(
{"--prompt-cache-all"},
"if specified, saves user input and generations to cache as well\n",
[](common_params & params) {
params.prompt_cache_all = true;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION}));
add_opt(common_arg(
{"--prompt-cache-ro"},
"if specified, uses the prompt cache but does not update it",
[](common_params & params) {
params.prompt_cache_ro = true;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION}));
add_opt(common_arg(
{"-r", "--reverse-prompt"}, "PROMPT",
"halt generation at PROMPT, return control in interactive mode\n",
[](common_params & params, const std::string & value) {
params.antiprompt.emplace_back(value);
}
- ).set_examples({LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_SERVER}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI, LLAMA_EXAMPLE_SERVER}));
add_opt(common_arg(
{"-sp", "--special"},
string_format("special tokens output enabled (default: %s)", params.special ? "true" : "false"),
[](common_params & params) {
params.special = true;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_SERVER}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI, LLAMA_EXAMPLE_SERVER}));
add_opt(common_arg(
{"-cnv", "--conversation"},
- "run in conversation mode:\n"
+ {"-no-cnv", "--no-conversation"},
+ "whether to run in conversation mode:\n"
"- does not print special tokens and suffix/prefix\n"
"- interactive mode is also enabled\n"
"(default: auto enabled if chat template is available)",
- [](common_params & params) {
- params.conversation_mode = COMMON_CONVERSATION_MODE_ENABLED;
+ [](common_params & params, bool value) {
+ params.conversation_mode = value ? COMMON_CONVERSATION_MODE_ENABLED : COMMON_CONVERSATION_MODE_DISABLED;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN}));
- add_opt(common_arg(
- {"-no-cnv", "--no-conversation"},
- "force disable conversation mode (default: false)",
- [](common_params & params) {
- params.conversation_mode = COMMON_CONVERSATION_MODE_DISABLED;
- }
- ).set_examples({LLAMA_EXAMPLE_MAIN}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI}));
add_opt(common_arg(
{"-st", "--single-turn"},
"run conversation for a single turn only, then exit when done\n"
@@ -1169,28 +1344,28 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
[](common_params & params) {
params.single_turn = true;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI}));
add_opt(common_arg(
{"-i", "--interactive"},
string_format("run in interactive mode (default: %s)", params.interactive ? "true" : "false"),
[](common_params & params) {
params.interactive = true;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION}));
add_opt(common_arg(
{"-if", "--interactive-first"},
string_format("run in interactive mode and wait for input right away (default: %s)", params.interactive_first ? "true" : "false"),
[](common_params & params) {
params.interactive_first = true;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION}));
add_opt(common_arg(
{"-mli", "--multiline-input"},
"allows you to write or paste multiple lines without ending each in '\\'",
[](common_params & params) {
params.multiline_input = true;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI}));
add_opt(common_arg(
{"--in-prefix-bos"},
"prefix BOS to user inputs, preceding the `--in-prefix` string",
@@ -1198,7 +1373,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.input_prefix_bos = true;
params.enable_chat_template = false;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION}));
add_opt(common_arg(
{"--in-prefix"}, "STRING",
"string to prefix user inputs with (default: empty)",
@@ -1206,7 +1381,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.input_prefix = value;
params.enable_chat_template = false;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION}));
add_opt(common_arg(
{"--in-suffix"}, "STRING",
"string to suffix after user inputs with (default: empty)",
@@ -1214,14 +1389,15 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.input_suffix = value;
params.enable_chat_template = false;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION}));
add_opt(common_arg(
+ {"--warmup"},
{"--no-warmup"},
- "skip warming up the model with an empty run",
- [](common_params & params) {
- params.warmup = false;
+ string_format("whether to perform warmup with an empty run (default: %s)", params.warmup ? "enabled" : "disabled"),
+ [](common_params & params, bool value) {
+ params.warmup = value;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_EMBEDDING, LLAMA_EXAMPLE_RETRIEVAL, LLAMA_EXAMPLE_PERPLEXITY}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_MTMD, LLAMA_EXAMPLE_EMBEDDING, LLAMA_EXAMPLE_RETRIEVAL, LLAMA_EXAMPLE_PERPLEXITY}));
add_opt(common_arg(
{"--spm-infill"},
string_format(
@@ -1249,7 +1425,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
}
).set_sparam());
add_opt(common_arg(
- {"--sampling-seq", "--sampler-seq"}, "SEQUENCE",
+ {"--sampler-seq", "--sampling-seq"}, "SEQUENCE",
string_format("simplified sequence for samplers that will be used (default: %s)", sampler_type_chars.c_str()),
[](common_params & params, const std::string & value) {
params.sampling.samplers = common_sampler_types_from_chars(value);
@@ -1278,7 +1454,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.sampling.top_k = value;
params.sampling.user_sampling_config |= common_params_sampling_config::COMMON_PARAMS_SAMPLING_CONFIG_TOP_K;
}
- ).set_sparam());
+ ).set_sparam().set_env("LLAMA_ARG_TOP_K"));
add_opt(common_arg(
{"--top-p"}, "N",
string_format("top-p sampling (default: %.1f, 1.0 = disabled)", (double)params.sampling.top_p),
@@ -1519,6 +1695,13 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.sampling.grammar = json_schema_to_grammar(json::parse(schema));
}
).set_sparam());
+ add_opt(common_arg(
+ {"-bs", "--backend-sampling"},
+ "enable backend sampling (experimental) (default: disabled)",
+ [](common_params & params) {
+ params.sampling.backend_sampling = true;
+ }
+ ).set_sparam().set_env("LLAMA_ARG_BACKEND_SAMPLING"));
add_opt(common_arg(
{"--pooling"}, "{none,mean,cls,last,rank}",
"pooling type for embeddings, use model default if unspecified",
@@ -1612,28 +1795,30 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
[](common_params & params, int value) {
params.grp_attn_n = value;
}
- ).set_env("LLAMA_ARG_GRP_ATTN_N").set_examples({LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_PASSKEY}));
+ ).set_env("LLAMA_ARG_GRP_ATTN_N").set_examples({LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_PASSKEY}));
add_opt(common_arg(
{"-gaw", "--grp-attn-w"}, "N",
string_format("group-attention width (default: %d)", params.grp_attn_w),
[](common_params & params, int value) {
params.grp_attn_w = value;
}
- ).set_env("LLAMA_ARG_GRP_ATTN_W").set_examples({LLAMA_EXAMPLE_MAIN}));
+ ).set_env("LLAMA_ARG_GRP_ATTN_W").set_examples({LLAMA_EXAMPLE_COMPLETION}));
add_opt(common_arg(
+ {"-kvo", "--kv-offload"},
{"-nkvo", "--no-kv-offload"},
- "disable KV offload",
- [](common_params & params) {
- params.no_kv_offload = true;
+ string_format("whether to enable KV cache offloading (default: %s)", params.no_kv_offload ? "disabled" : "enabled"),
+ [](common_params & params, bool value) {
+ params.no_kv_offload = !value;
}
- ).set_env("LLAMA_ARG_NO_KV_OFFLOAD"));
+ ).set_env("LLAMA_ARG_KV_OFFLOAD"));
add_opt(common_arg(
+ {"--repack"},
{"-nr", "--no-repack"},
- "disable weight repacking",
- [](common_params & params) {
- params.no_extra_bufts = true;
+ string_format("whether to enable weight repacking (default: %s)", params.no_extra_bufts ? "disabled" : "enabled"),
+ [](common_params & params, bool value) {
+ params.no_extra_bufts = !value;
}
- ).set_env("LLAMA_ARG_NO_REPACK"));
+ ).set_env("LLAMA_ARG_REPACK"));
add_opt(common_arg(
{"--no-host"},
"bypass host buffer allowing extra buffers to be used",
@@ -1746,13 +1931,27 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
LOG_WRN("DEPRECATED: --defrag-thold is deprecated and no longer necessary to specify\n");
}
).set_env("LLAMA_ARG_DEFRAG_THOLD"));
- add_opt(common_arg(
- {"-np", "--parallel"}, "N",
- string_format("number of parallel sequences to decode (default: %d)", params.n_parallel),
- [](common_params & params, int value) {
- params.n_parallel = value;
- }
- ).set_env("LLAMA_ARG_N_PARALLEL"));
+ if (ex == LLAMA_EXAMPLE_SERVER) {
+ // this is to make sure this option appears in the server-specific section of the help message
+ add_opt(common_arg(
+ {"-np", "--parallel"}, "N",
+ string_format("number of server slots (default: %d, -1 = auto)", params.n_parallel),
+ [](common_params & params, int value) {
+ if (value == 0) {
+ throw std::invalid_argument("error: invalid value for n_parallel\n");
+ }
+ params.n_parallel = value;
+ }
+ ).set_env("LLAMA_ARG_N_PARALLEL").set_examples({LLAMA_EXAMPLE_SERVER}));
+ } else {
+ add_opt(common_arg(
+ {"-np", "--parallel"}, "N",
+ string_format("number of parallel sequences to decode (default: %d)", params.n_parallel),
+ [](common_params & params, int value) {
+ params.n_parallel = value;
+ }
+ ).set_env("LLAMA_ARG_N_PARALLEL"));
+ }
add_opt(common_arg(
{"-ns", "--sequences"}, "N",
string_format("number of sequences to decode (default: %d)", params.n_sequences),
@@ -1762,20 +1961,14 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
).set_examples({LLAMA_EXAMPLE_PARALLEL}));
add_opt(common_arg(
{"-cb", "--cont-batching"},
- string_format("enable continuous batching (a.k.a dynamic batching) (default: %s)", params.cont_batching ? "enabled" : "disabled"),
- [](common_params & params) {
- params.cont_batching = true;
+ {"-nocb", "--no-cont-batching"},
+ string_format("whether to enable continuous batching (a.k.a dynamic batching) (default: %s)", params.cont_batching ? "enabled" : "disabled"),
+ [](common_params & params, bool value) {
+ params.cont_batching = value;
}
).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_CONT_BATCHING"));
add_opt(common_arg(
- {"-nocb", "--no-cont-batching"},
- "disable continuous batching",
- [](common_params & params) {
- params.cont_batching = false;
- }
- ).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_NO_CONT_BATCHING"));
- add_opt(common_arg(
- {"--mmproj"}, "FILE",
+ {"-mm", "--mmproj"}, "FILE",
"path to a multimodal projector file. see tools/mtmd/README.md\n"
"note: if -hf is used, this argument can be omitted",
[](common_params & params, const std::string & value) {
@@ -1783,33 +1976,37 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
}
).set_examples(mmproj_examples).set_env("LLAMA_ARG_MMPROJ"));
add_opt(common_arg(
- {"--mmproj-url"}, "URL",
+ {"-mmu", "--mmproj-url"}, "URL",
"URL to a multimodal projector file. see tools/mtmd/README.md",
[](common_params & params, const std::string & value) {
params.mmproj.url = value;
}
).set_examples(mmproj_examples).set_env("LLAMA_ARG_MMPROJ_URL"));
add_opt(common_arg(
- {"--no-mmproj"},
- "explicitly disable multimodal projector, useful when using -hf",
- [](common_params & params) {
- params.no_mmproj = true;
+ {"--mmproj-auto"},
+ {"--no-mmproj", "--no-mmproj-auto"},
+ string_format("whether to use multimodal projector file (if available), useful when using -hf (default: %s)", params.no_mmproj ? "disabled" : "enabled"),
+ [](common_params & params, bool value) {
+ params.no_mmproj = !value;
}
- ).set_examples(mmproj_examples).set_env("LLAMA_ARG_NO_MMPROJ"));
+ ).set_examples(mmproj_examples).set_env("LLAMA_ARG_MMPROJ_AUTO"));
add_opt(common_arg(
+ {"--mmproj-offload"},
{"--no-mmproj-offload"},
- "do not offload multimodal projector to GPU",
- [](common_params & params) {
- params.mmproj_use_gpu = false;
+ string_format("whether to enable GPU offloading for multimodal projector (default: %s)", params.mmproj_use_gpu ? "enabled" : "disabled"),
+ [](common_params & params, bool value) {
+ params.mmproj_use_gpu = value;
}
- ).set_examples(mmproj_examples).set_env("LLAMA_ARG_NO_MMPROJ_OFFLOAD"));
+ ).set_examples(mmproj_examples).set_env("LLAMA_ARG_MMPROJ_OFFLOAD"));
add_opt(common_arg(
{"--image", "--audio"}, "FILE",
- "path to an image or audio file. use with multimodal models, can be repeated if you have multiple files\n",
+ "path to an image or audio file. use with multimodal models, use comma-separated values for multiple files\n",
[](common_params & params, const std::string & value) {
- params.image.emplace_back(value);
+ for (const auto & item : string_split(value, ',')) {
+ params.image.emplace_back(item);
+ }
}
- ).set_examples({LLAMA_EXAMPLE_MTMD}));
+ ).set_examples({LLAMA_EXAMPLE_MTMD, LLAMA_EXAMPLE_CLI}));
add_opt(common_arg(
{"--image-min-tokens"}, "N",
"minimum number of tokens each image can take, only used by vision models with dynamic resolution (default: read from model)",
@@ -1827,7 +2024,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
if (llama_supports_rpc()) {
add_opt(common_arg(
{"--rpc"}, "SERVERS",
- "comma separated list of RPC servers",
+ "comma separated list of RPC servers (host:port)",
[](common_params & params, const std::string & value) {
add_rpc_devices(value);
GGML_UNUSED(params);
@@ -1842,12 +2039,13 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
}
).set_env("LLAMA_ARG_MLOCK"));
add_opt(common_arg(
+ {"--mmap"},
{"--no-mmap"},
- "do not memory-map model (slower load but may reduce pageouts if not using mlock)",
- [](common_params & params) {
- params.use_mmap = false;
+ string_format("whether to memory-map model (if disabled, slower load but may reduce pageouts if not using mlock) (default: %s)", params.use_mmap ? "enabled" : "disabled"),
+ [](common_params & params, bool value) {
+ params.use_mmap = value;
}
- ).set_env("LLAMA_ARG_NO_MMAP"));
+ ).set_env("LLAMA_ARG_MMAP"));
add_opt(common_arg(
{"--numa"}, "TYPE",
"attempt optimizations that help on some NUMA systems\n"
@@ -1892,26 +2090,26 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
}
));
add_opt(common_arg(
- {"--override-tensor", "-ot"}, "=,...",
+ {"-ot", "--override-tensor"}, "=,...",
"override tensor buffer type", [](common_params & params, const std::string & value) {
parse_tensor_buffer_overrides(value, params.tensor_buft_overrides);
}
- ));
+ ).set_env("LLAMA_ARG_OVERRIDE_TENSOR"));
add_opt(common_arg(
- {"--override-tensor-draft", "-otd"}, "=,...",
+ {"-otd", "--override-tensor-draft"}, "=,...",
"override tensor buffer type for draft model", [](common_params & params, const std::string & value) {
parse_tensor_buffer_overrides(value, params.speculative.tensor_buft_overrides);
}
- ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER}));
+ ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}));
add_opt(common_arg(
- {"--cpu-moe", "-cmoe"},
+ {"-cmoe", "--cpu-moe"},
"keep all Mixture of Experts (MoE) weights in the CPU",
[](common_params & params) {
params.tensor_buft_overrides.push_back(llm_ffn_exps_cpu_override());
}
).set_env("LLAMA_ARG_CPU_MOE"));
add_opt(common_arg(
- {"--n-cpu-moe", "-ncmoe"}, "N",
+ {"-ncmoe", "--n-cpu-moe"}, "N",
"keep the Mixture of Experts (MoE) weights of the first N layers in the CPU",
[](common_params & params, int value) {
if (value < 0) {
@@ -1926,14 +2124,14 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
}
).set_env("LLAMA_ARG_N_CPU_MOE"));
add_opt(common_arg(
- {"--cpu-moe-draft", "-cmoed"},
+ {"-cmoed", "--cpu-moe-draft"},
"keep all Mixture of Experts (MoE) weights in the CPU for the draft model",
[](common_params & params) {
params.speculative.tensor_buft_overrides.push_back(llm_ffn_exps_cpu_override());
}
- ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_CPU_MOE_DRAFT"));
+ ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}).set_env("LLAMA_ARG_CPU_MOE_DRAFT"));
add_opt(common_arg(
- {"--n-cpu-moe-draft", "-ncmoed"}, "N",
+ {"-ncmoed", "--n-cpu-moe-draft"}, "N",
"keep the Mixture of Experts (MoE) weights of the first N layers in the CPU for the draft model",
[](common_params & params, int value) {
if (value < 0) {
@@ -1945,12 +2143,19 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.speculative.tensor_buft_overrides.push_back({buft_overrides_draft.back().c_str(), ggml_backend_cpu_buffer_type()});
}
}
- ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_N_CPU_MOE_DRAFT"));
+ ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}).set_env("LLAMA_ARG_N_CPU_MOE_DRAFT"));
+ GGML_ASSERT(params.n_gpu_layers < 0); // string_format would need to be extended for a default >= 0
add_opt(common_arg(
{"-ngl", "--gpu-layers", "--n-gpu-layers"}, "N",
- string_format("max. number of layers to store in VRAM (default: %d)", params.n_gpu_layers),
- [](common_params & params, int value) {
- params.n_gpu_layers = value;
+ string_format("max. number of layers to store in VRAM, either an exact number, 'auto', or 'all' (default: %s)", params.n_gpu_layers == -1 ? "auto" : "all"),
+ [](common_params & params, const std::string & value) {
+ if (value == "auto") {
+ params.n_gpu_layers = -1;
+ } else if (value == "all") {
+ params.n_gpu_layers = -2;
+ } else {
+ params.n_gpu_layers = std::stoi(value);
+ }
if (!llama_supports_gpu_offload()) {
fprintf(stderr, "warning: no usable GPU found, --gpu-layers option will be ignored\n");
fprintf(stderr, "warning: one possible reason is that llama.cpp was compiled without GPU support\n");
@@ -2017,6 +2222,34 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
}
}
).set_env("LLAMA_ARG_MAIN_GPU"));
+ add_opt(common_arg(
+ { "-fit", "--fit" }, "[on|off]",
+ string_format("whether to adjust unset arguments to fit in device memory ('on' or 'off', default: '%s')", params.fit_params ? "on" : "off"),
+ [](common_params & params, const std::string & value) {
+ if (is_truthy(value)) {
+ params.fit_params = true;
+ } else if (is_falsey(value)) {
+ params.fit_params = false;
+ } else {
+ throw std::runtime_error(
+ string_format("error: unkown value for --fit: '%s'\n", value.c_str()));
+ }
+ }
+ ).set_env("LLAMA_ARG_FIT"));
+ add_opt(common_arg(
+ { "-fitt", "--fit-target" }, "MiB",
+ string_format("target margin per device for --fit option, default: %zu", params.fit_params_target/(1024*1024)),
+ [](common_params & params, int value) {
+ params.fit_params_target = value * size_t(1024*1024);
+ }
+ ).set_env("LLAMA_ARG_FIT_TARGET"));
+ add_opt(common_arg(
+ { "-fitc", "--fit-ctx" }, "N",
+ string_format("minimum ctx size that can be set by --fit option, default: %" PRIu32, params.fit_params_min_ctx),
+ [](common_params & params, int value) {
+ params.fit_params_min_ctx = value;
+ }
+ ).set_env("LLAMA_ARG_FIT_CTX"));
add_opt(common_arg(
{"--check-tensors"},
string_format("check model tensor data for invalid values (default: %s)", params.check_tensors ? "true" : "false"),
@@ -2025,51 +2258,96 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
}
));
add_opt(common_arg(
- {"--override-kv"}, "KEY=TYPE:VALUE",
- "advanced option to override model metadata by key. may be specified multiple times.\n"
- "types: int, float, bool, str. example: --override-kv tokenizer.ggml.add_bos_token=bool:false",
+ {"--override-kv"}, "KEY=TYPE:VALUE,...",
+ "advanced option to override model metadata by key. to specify multiple overrides, either use comma-separated or repeat this argument.\n"
+ "types: int, float, bool, str. example: --override-kv tokenizer.ggml.add_bos_token=bool:false,tokenizer.ggml.add_eos_token=bool:false",
[](common_params & params, const std::string & value) {
- if (!string_parse_kv_override(value.c_str(), params.kv_overrides)) {
- throw std::runtime_error(string_format("error: Invalid type for KV override: %s\n", value.c_str()));
+ std::vector kv_overrides;
+
+ std::string current;
+ bool escaping = false;
+
+ for (const char c : value) {
+ if (escaping) {
+ current.push_back(c);
+ escaping = false;
+ } else if (c == '\\') {
+ escaping = true;
+ } else if (c == ',') {
+ kv_overrides.push_back(current);
+ current.clear();
+ } else {
+ current.push_back(c);
+ }
+ }
+
+ if (escaping) {
+ current.push_back('\\');
+ }
+
+ kv_overrides.push_back(current);
+
+ for (const auto & kv_override : kv_overrides) {
+ if (!string_parse_kv_override(kv_override.c_str(), params.kv_overrides)) {
+ throw std::runtime_error(string_format("error: Invalid type for KV override: %s\n", kv_override.c_str()));
+ }
}
}
));
add_opt(common_arg(
+ {"--op-offload"},
{"--no-op-offload"},
- string_format("disable offloading host tensor operations to device (default: %s)", params.no_op_offload ? "true" : "false"),
- [](common_params & params) {
- params.no_op_offload = true;
+ string_format("whether to offload host tensor operations to device (default: %s)", params.no_op_offload ? "false" : "true"),
+ [](common_params & params, bool value) {
+ params.no_op_offload = !value;
}
));
add_opt(common_arg(
{"--lora"}, "FNAME",
- "path to LoRA adapter (can be repeated to use multiple adapters)",
+ "path to LoRA adapter (use comma-separated values to load multiple adapters)",
[](common_params & params, const std::string & value) {
- params.lora_adapters.push_back({ std::string(value), 1.0, "", "", nullptr });
+ for (const auto & item : string_split(value, ',')) {
+ params.lora_adapters.push_back({ item, 1.0, "", "", nullptr });
+ }
}
// we define this arg on both COMMON and EXPORT_LORA, so when showing help message of export-lora, it will be categorized as "example-specific" arg
).set_examples({LLAMA_EXAMPLE_COMMON, LLAMA_EXAMPLE_EXPORT_LORA}));
add_opt(common_arg(
- {"--lora-scaled"}, "FNAME", "SCALE",
- "path to LoRA adapter with user defined scaling (can be repeated to use multiple adapters)",
- [](common_params & params, const std::string & fname, const std::string & scale) {
- params.lora_adapters.push_back({ fname, std::stof(scale), "", "", nullptr });
+ {"--lora-scaled"}, "FNAME:SCALE,...",
+ "path to LoRA adapter with user defined scaling (format: FNAME:SCALE,...)\n"
+ "note: use comma-separated values",
+ [](common_params & params, const std::string & value) {
+ for (const auto & item : string_split(value, ',')) {
+ auto parts = string_split(item, ':');
+ if (parts.size() != 2) {
+ throw std::invalid_argument("lora-scaled format: FNAME:SCALE");
+ }
+ params.lora_adapters.push_back({ parts[0], std::stof(parts[1]), "", "", nullptr });
+ }
}
// we define this arg on both COMMON and EXPORT_LORA, so when showing help message of export-lora, it will be categorized as "example-specific" arg
).set_examples({LLAMA_EXAMPLE_COMMON, LLAMA_EXAMPLE_EXPORT_LORA}));
add_opt(common_arg(
{"--control-vector"}, "FNAME",
- "add a control vector\nnote: this argument can be repeated to add multiple control vectors",
+ "add a control vector\nnote: use comma-separated values to add multiple control vectors",
[](common_params & params, const std::string & value) {
- params.control_vectors.push_back({ 1.0f, value, });
+ for (const auto & item : string_split(value, ',')) {
+ params.control_vectors.push_back({ 1.0f, item, });
+ }
}
));
add_opt(common_arg(
- {"--control-vector-scaled"}, "FNAME", "SCALE",
+ {"--control-vector-scaled"}, "FNAME:SCALE,...",
"add a control vector with user defined scaling SCALE\n"
- "note: this argument can be repeated to add multiple scaled control vectors",
- [](common_params & params, const std::string & fname, const std::string & scale) {
- params.control_vectors.push_back({ std::stof(scale), fname });
+ "note: use comma-separated values (format: FNAME:SCALE,...)",
+ [](common_params & params, const std::string & value) {
+ for (const auto & item : string_split(value, ',')) {
+ auto parts = string_split(item, ':');
+ if (parts.size() != 2) {
+ throw std::invalid_argument("control-vector-scaled format: FNAME:SCALE");
+ }
+ params.control_vectors.push_back({ std::stof(parts[1]), parts[0] });
+ }
}
));
add_opt(common_arg(
@@ -2090,11 +2368,8 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
add_opt(common_arg(
{"-m", "--model"}, "FNAME",
ex == LLAMA_EXAMPLE_EXPORT_LORA
- ? std::string("model path from which to load base model")
- : string_format(
- "model path (default: `models/$filename` with filename from `--hf-file` "
- "or `--model-url` if set, otherwise %s)", DEFAULT_MODEL_PATH
- ),
+ ? "model path from which to load base model"
+ : "model path to load",
[](common_params & params, const std::string & value) {
params.model.path = value;
}
@@ -2162,13 +2437,15 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
).set_env("HF_TOKEN"));
add_opt(common_arg(
{"--context-file"}, "FNAME",
- "file to load context from (repeat to specify multiple files)",
+ "file to load context from (use comma-separated values to specify multiple files)",
[](common_params & params, const std::string & value) {
- std::ifstream file(value, std::ios::binary);
- if (!file) {
- throw std::runtime_error(string_format("error: failed to open file '%s'\n", value.c_str()));
+ for (const auto & item : string_split(value, ',')) {
+ std::ifstream file(item, std::ios::binary);
+ if (!file) {
+ throw std::runtime_error(string_format("error: failed to open file '%s'\n", item.c_str()));
+ }
+ params.context_files.push_back(item);
}
- params.context_files.push_back(value);
}
).set_examples({LLAMA_EXAMPLE_RETRIEVAL}));
add_opt(common_arg(
@@ -2237,10 +2514,11 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
}
).set_examples({LLAMA_EXAMPLE_IMATRIX}));
add_opt(common_arg(
+ {"--ppl"},
{"--no-ppl"},
- string_format("do not compute perplexity (default: %s)", params.compute_ppl ? "true" : "false"),
- [](common_params & params) {
- params.compute_ppl = false;
+ string_format("whether to compute perplexity (default: %s)", params.compute_ppl ? "true" : "false"),
+ [](common_params & params, bool value) {
+ params.compute_ppl = value;
}
).set_examples({LLAMA_EXAMPLE_IMATRIX}));
add_opt(common_arg(
@@ -2359,12 +2637,27 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
}
).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_API_PREFIX"));
add_opt(common_arg(
- {"--no-webui"},
- string_format("Disable the Web UI (default: %s)", params.webui ? "enabled" : "disabled"),
- [](common_params & params) {
- params.webui = false;
+ {"--webui-config"}, "JSON",
+ "JSON that provides default WebUI settings (overrides WebUI defaults)",
+ [](common_params & params, const std::string & value) {
+ params.webui_config_json = value;
}
- ).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_NO_WEBUI"));
+ ).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_WEBUI_CONFIG"));
+ add_opt(common_arg(
+ {"--webui-config-file"}, "PATH",
+ "JSON file that provides default WebUI settings (overrides WebUI defaults)",
+ [](common_params & params, const std::string & value) {
+ params.webui_config_json = read_file(value);
+ }
+ ).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_WEBUI_CONFIG_FILE"));
+ add_opt(common_arg(
+ {"--webui"},
+ {"--no-webui"},
+ string_format("whether to enable the Web UI (default: %s)", params.webui ? "enabled" : "disabled"),
+ [](common_params & params, bool value) {
+ params.webui = value;
+ }
+ ).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_WEBUI"));
add_opt(common_arg(
{"--embedding", "--embeddings"},
string_format("restrict to only support embedding use case; use only with dedicated embedding models (default: %s)", params.embedding ? "enabled" : "disabled"),
@@ -2373,7 +2666,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
}
).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_EMBEDDINGS"));
add_opt(common_arg(
- {"--reranking", "--rerank"},
+ {"--rerank", "--reranking"},
string_format("enable reranking endpoint on server (default: %s)", "disabled"),
[](common_params & params) {
params.embedding = true;
@@ -2427,7 +2720,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.default_template_kwargs[item.key()] = item.value().dump();
}
}
- ).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_CHAT_TEMPLATE_KWARGS"));
+ ).set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}).set_env("LLAMA_CHAT_TEMPLATE_KWARGS"));
add_opt(common_arg(
{"-to", "--timeout"}, "N",
string_format("server read/write timeout in seconds (default: %d)", params.timeout_read),
@@ -2469,23 +2762,20 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_ENDPOINT_PROPS"));
add_opt(common_arg(
{"--slots"},
- string_format("enable slots monitoring endpoint (default: %s)", params.endpoint_slots ? "enabled" : "disabled"),
- [](common_params & params) {
- params.endpoint_slots = true;
+ {"--no-slots"},
+ string_format("expose slots monitoring endpoint (default: %s)", params.endpoint_slots ? "enabled" : "disabled"),
+ [](common_params & params, bool value) {
+ params.endpoint_slots = value;
}
).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_ENDPOINT_SLOTS"));
- add_opt(common_arg(
- {"--no-slots"},
- "disables slots monitoring endpoint",
- [](common_params & params) {
- params.endpoint_slots = false;
- }
- ).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_NO_ENDPOINT_SLOTS"));
add_opt(common_arg(
{"--slot-save-path"}, "PATH",
"path to save slot kv cache (default: disabled)",
[](common_params & params, const std::string & value) {
params.slot_save_path = value;
+ if (!fs_is_directory(params.slot_save_path)) {
+ throw std::invalid_argument("not a directory: " + value);
+ }
// if doesn't end with DIRECTORY_SEPARATOR, add it
if (!params.slot_save_path.empty() && params.slot_save_path[params.slot_save_path.size() - 1] != DIRECTORY_SEPARATOR) {
params.slot_save_path += DIRECTORY_SEPARATOR;
@@ -2493,19 +2783,56 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
}
).set_examples({LLAMA_EXAMPLE_SERVER}));
add_opt(common_arg(
- {"--jinja"},
- string_format("use jinja template for chat (default: %s)\n", params.use_jinja ? "enabled" : "disabled"),
- [](common_params & params) {
- params.use_jinja = true;
+ {"--media-path"}, "PATH",
+ "directory for loading local media files; files can be accessed via file:// URLs using relative paths (default: disabled)",
+ [](common_params & params, const std::string & value) {
+ params.media_path = value;
+ if (!fs_is_directory(params.media_path)) {
+ throw std::invalid_argument("not a directory: " + value);
+ }
+ // if doesn't end with DIRECTORY_SEPARATOR, add it
+ if (!params.media_path.empty() && params.media_path[params.media_path.size() - 1] != DIRECTORY_SEPARATOR) {
+ params.media_path += DIRECTORY_SEPARATOR;
+ }
}
- ).set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_MTMD}).set_env("LLAMA_ARG_JINJA"));
+ ).set_examples({LLAMA_EXAMPLE_SERVER}));
add_opt(common_arg(
- {"--no-jinja"},
- string_format("disable jinja template for chat (default: %s)\n", params.use_jinja ? "enabled" : "disabled"),
- [](common_params & params) {
- params.use_jinja = false;
+ {"--models-dir"}, "PATH",
+ "directory containing models for the router server (default: disabled)",
+ [](common_params & params, const std::string & value) {
+ params.models_dir = value;
}
- ).set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_MTMD}).set_env("LLAMA_ARG_NO_JINJA"));
+ ).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_MODELS_DIR"));
+ add_opt(common_arg(
+ {"--models-preset"}, "PATH",
+ "path to INI file containing model presets for the router server (default: disabled)",
+ [](common_params & params, const std::string & value) {
+ params.models_preset = value;
+ }
+ ).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_MODELS_PRESET"));
+ add_opt(common_arg(
+ {"--models-max"}, "N",
+ string_format("for router server, maximum number of models to load simultaneously (default: %d, 0 = unlimited)", params.models_max),
+ [](common_params & params, int value) {
+ params.models_max = value;
+ }
+ ).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_MODELS_MAX"));
+ add_opt(common_arg(
+ {"--models-autoload"},
+ {"--no-models-autoload"},
+ string_format("for router server, whether to automatically load models (default: %s)", params.models_autoload ? "enabled" : "disabled"),
+ [](common_params & params, bool value) {
+ params.models_autoload = value;
+ }
+ ).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_MODELS_AUTOLOAD"));
+ add_opt(common_arg(
+ {"--jinja"},
+ {"--no-jinja"},
+ string_format("whether to use jinja template engine for chat (default: %s)", params.use_jinja ? "enabled" : "disabled"),
+ [](common_params & params, bool value) {
+ params.use_jinja = value;
+ }
+ ).set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI, LLAMA_EXAMPLE_MTMD}).set_env("LLAMA_ARG_JINJA"));
add_opt(common_arg(
{"--reasoning-format"}, "FORMAT",
"controls whether thought tags are allowed and/or extracted from the response, and in which format they're returned; one of:\n"
@@ -2516,7 +2843,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
[](common_params & params, const std::string & value) {
params.reasoning_format = common_reasoning_format_from_name(value);
}
- ).set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_MAIN}).set_env("LLAMA_ARG_THINK"));
+ ).set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI}).set_env("LLAMA_ARG_THINK"));
add_opt(common_arg(
{"--reasoning-budget"}, "N",
"controls the amount of thinking allowed; currently only one of: -1 for unrestricted thinking budget, or 0 to disable thinking (default: -1)",
@@ -2524,7 +2851,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
if (value != 0 && value != -1) { throw std::invalid_argument("invalid value"); }
params.reasoning_budget = value;
}
- ).set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_MAIN}).set_env("LLAMA_ARG_THINK_BUDGET"));
+ ).set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI}).set_env("LLAMA_ARG_THINK_BUDGET"));
add_opt(common_arg(
{"--chat-template"}, "JINJA_TEMPLATE",
string_format(
@@ -2536,7 +2863,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
[](common_params & params, const std::string & value) {
params.chat_template = value;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_MTMD}).set_env("LLAMA_ARG_CHAT_TEMPLATE"));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_MTMD}).set_env("LLAMA_ARG_CHAT_TEMPLATE"));
add_opt(common_arg(
{"--chat-template-file"}, "JINJA_TEMPLATE_FILE",
string_format(
@@ -2548,17 +2875,18 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
[](common_params & params, const std::string & value) {
params.chat_template = read_file(value);
}
- ).set_examples({LLAMA_EXAMPLE_MAIN, LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_CHAT_TEMPLATE_FILE"));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI, LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_CHAT_TEMPLATE_FILE"));
add_opt(common_arg(
+ {"--prefill-assistant"},
{"--no-prefill-assistant"},
string_format(
"whether to prefill the assistant's response if the last message is an assistant message (default: prefill enabled)\n"
"when this flag is set, if the last message is an assistant message then it will be treated as a full message and not prefilled\n"
),
- [](common_params & params) {
- params.prefill_assistant = false;
+ [](common_params & params, bool value) {
+ params.prefill_assistant = value;
}
- ).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_NO_PREFILL_ASSISTANT"));
+ ).set_examples({LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_PREFILL_ASSISTANT"));
add_opt(common_arg(
{"-sps", "--slot-prompt-similarity"}, "SIMILARITY",
string_format("how much the prompt of a request must match the prompt of a slot in order to use that slot (default: %.2f, 0.0 = disabled)\n", params.slot_prompt_similarity),
@@ -2573,13 +2901,23 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.lora_init_without_apply = true;
}
).set_examples({LLAMA_EXAMPLE_SERVER}));
+ add_opt(common_arg(
+ {"--sleep-idle-seconds"}, "SECONDS",
+ string_format("number of seconds of idleness after which the server will sleep (default: %d; -1 = disabled)", params.sleep_idle_seconds),
+ [](common_params & params, int value) {
+ if (value == 0 || value < -1) {
+ throw std::invalid_argument("invalid value: cannot be 0 or less than -1");
+ }
+ params.sleep_idle_seconds = value;
+ }
+ ).set_examples({LLAMA_EXAMPLE_SERVER}));
add_opt(common_arg(
{"--simple-io"},
"use basic IO for better compatibility in subprocesses and limited consoles",
[](common_params & params) {
params.simple_io = true;
}
- ).set_examples({LLAMA_EXAMPLE_MAIN}));
+ ).set_examples({LLAMA_EXAMPLE_COMPLETION, LLAMA_EXAMPLE_CLI}));
add_opt(common_arg(
{"--positive-file"}, "FNAME",
string_format("positive prompts file, one prompt per line (default: '%s')", params.cvector_positive_file.c_str()),
@@ -2653,7 +2991,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
common_log_set_colors(common_log_main(), LOG_COLORS_AUTO);
} else {
throw std::invalid_argument(
- string_format("error: unkown value for --log-colors: '%s'\n", value.c_str()));
+ string_format("error: unknown value for --log-colors: '%s'\n", value.c_str()));
}
}
).set_env("LLAMA_LOG_COLORS"));
@@ -2662,7 +3000,6 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
"Set verbosity level to infinity (i.e. log all messages, useful for debugging)",
[](common_params & params) {
params.verbosity = INT_MAX;
- common_log_set_verbosity_thold(INT_MAX);
}
));
add_opt(common_arg(
@@ -2683,7 +3020,6 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
"(default: %d)\n", params.verbosity),
[](common_params & params, int value) {
params.verbosity = value;
- common_log_set_verbosity_thold(value);
}
).set_env("LLAMA_LOG_VERBOSITY"));
add_opt(common_arg(
@@ -2811,19 +3147,19 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
}
).set_examples({LLAMA_EXAMPLE_SPECULATIVE}));
add_opt(common_arg(
- {"--draft-max", "--draft", "--draft-n"}, "N",
+ {"--draft", "--draft-n", "--draft-max"}, "N",
string_format("number of tokens to draft for speculative decoding (default: %d)", params.speculative.n_max),
[](common_params & params, int value) {
params.speculative.n_max = value;
}
- ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_LOOKUP, LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_DRAFT_MAX"));
+ ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_LOOKUP, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}).set_env("LLAMA_ARG_DRAFT_MAX"));
add_opt(common_arg(
{"--draft-min", "--draft-n-min"}, "N",
string_format("minimum number of draft tokens to use for speculative decoding (default: %d)", params.speculative.n_min),
[](common_params & params, int value) {
params.speculative.n_min = value;
}
- ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_LOOKUP, LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_DRAFT_MIN"));
+ ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_LOOKUP, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}).set_env("LLAMA_ARG_DRAFT_MIN"));
add_opt(common_arg(
{"--draft-p-split"}, "P",
string_format("speculative decoding split probability (default: %.1f)", (double)params.speculative.p_split),
@@ -2837,14 +3173,14 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
[](common_params & params, const std::string & value) {
params.speculative.p_min = std::stof(value);
}
- ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_DRAFT_P_MIN"));
+ ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}).set_env("LLAMA_ARG_DRAFT_P_MIN"));
add_opt(common_arg(
{"-cd", "--ctx-size-draft"}, "N",
string_format("size of the prompt context for the draft model (default: %d, 0 = loaded from model)", params.speculative.n_ctx),
[](common_params & params, int value) {
params.speculative.n_ctx = value;
}
- ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_CTX_SIZE_DRAFT"));
+ ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}).set_env("LLAMA_ARG_CTX_SIZE_DRAFT"));
add_opt(common_arg(
{"-devd", "--device-draft"}, "",
"comma-separated list of devices to use for offloading the draft model (none = don't offload)\n"
@@ -2852,33 +3188,41 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
[](common_params & params, const std::string & value) {
params.speculative.devices = parse_device_list(value);
}
- ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER}));
+ ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}));
+ GGML_ASSERT(params.speculative.n_gpu_layers < 0); // string_format would need to be extended for a default >= 0
add_opt(common_arg(
{"-ngld", "--gpu-layers-draft", "--n-gpu-layers-draft"}, "N",
- "number of layers to store in VRAM for the draft model",
- [](common_params & params, int value) {
- params.speculative.n_gpu_layers = value;
+ string_format("max. number of draft model layers to store in VRAM, either an exact number, 'auto', or 'all' (default: %s)",
+ params.speculative.n_gpu_layers == -1 ? "auto" : "all"),
+ [](common_params & params, const std::string & value) {
+ if (value == "auto") {
+ params.speculative.n_gpu_layers = -1;
+ } else if (value == "all") {
+ params.speculative.n_gpu_layers = -2;
+ } else {
+ params.speculative.n_gpu_layers = std::stoi(value);
+ }
if (!llama_supports_gpu_offload()) {
fprintf(stderr, "warning: no usable GPU found, --gpu-layers-draft option will be ignored\n");
fprintf(stderr, "warning: one possible reason is that llama.cpp was compiled without GPU support\n");
fprintf(stderr, "warning: consult docs/build.md for compilation instructions\n");
}
}
- ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_N_GPU_LAYERS_DRAFT"));
+ ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}).set_env("LLAMA_ARG_N_GPU_LAYERS_DRAFT"));
add_opt(common_arg(
{"-md", "--model-draft"}, "FNAME",
"draft model for speculative decoding (default: unused)",
[](common_params & params, const std::string & value) {
params.speculative.model.path = value;
}
- ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_MODEL_DRAFT"));
+ ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}).set_env("LLAMA_ARG_MODEL_DRAFT"));
add_opt(common_arg(
{"--spec-replace"}, "TARGET", "DRAFT",
"translate the string in TARGET into DRAFT if the draft model and main model are not compatible",
[](common_params & params, const std::string & tgt, const std::string & dft) {
params.speculative.replacements.push_back({ tgt, dft });
}
- ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER}));
+ ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}));
add_opt(common_arg(
{"-ctkd", "--cache-type-k-draft"}, "TYPE",
string_format(
@@ -3142,7 +3486,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.use_jinja = true;
//params.default_template_kwargs["reasoning_effort"] = "\"high\"";
}
- ).set_examples({LLAMA_EXAMPLE_SERVER}));
+ ).set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}));
add_opt(common_arg(
{"--gpt-oss-120b-default"},
@@ -3161,7 +3505,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.use_jinja = true;
//params.default_template_kwargs["reasoning_effort"] = "\"high\"";
}
- ).set_examples({LLAMA_EXAMPLE_SERVER}));
+ ).set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}));
add_opt(common_arg(
{"--vision-gemma-4b-default"},
@@ -3172,7 +3516,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.n_ctx = 0;
params.use_jinja = true;
}
- ).set_examples({LLAMA_EXAMPLE_SERVER}));
+ ).set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}));
add_opt(common_arg(
{"--vision-gemma-12b-default"},
@@ -3183,7 +3527,28 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.n_ctx = 0;
params.use_jinja = true;
}
- ).set_examples({LLAMA_EXAMPLE_SERVER}));
+ ).set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}));
return ctx_arg;
}
+
+void common_params_add_preset_options(std::vector & args) {
+ // arguments below won't be treated as CLI args, only preset options
+ args.push_back(common_arg(
+ {"load-on-startup"}, "NAME",
+ "in server router mode, autoload this model on startup",
+ [](common_params &, const std::string &) { /* unused */ }
+ ).set_env(COMMON_ARG_PRESET_LOAD_ON_STARTUP).set_preset_only());
+
+ args.push_back(common_arg(
+ {"stop-timeout"}, "SECONDS",
+ "in server router mode, force-kill model instance after this many seconds of graceful shutdown",
+ [](common_params &, int) { /* unused */ }
+ ).set_env(COMMON_ARG_PRESET_STOP_TIMEOUT).set_preset_only());
+
+ // args.push_back(common_arg(
+ // {"pin"},
+ // "in server router mode, do not unload this model if models_max is exceeded",
+ // [](common_params &) { /* unused */ }
+ // ).set_preset_only());
+}
diff --git a/common/arg.h b/common/arg.h
index 7ab7e2cea4..a1b6a14e67 100644
--- a/common/arg.h
+++ b/common/arg.h
@@ -3,8 +3,14 @@
#include "common.h"
#include
+#include