Packages
llama_cpp_ex
0.8.5
0.8.36
0.8.35
0.8.34
0.8.33
0.8.32
0.8.31
0.8.28
0.8.27
0.8.26
0.8.25
0.8.24
0.8.23
0.8.22
0.8.21
0.8.20
0.8.19
0.8.18
0.8.17
0.8.16
0.8.15
0.8.14
0.8.13
0.8.12
0.8.11
0.8.10
0.8.9
0.8.8
0.8.7
0.8.6
0.8.5
0.8.4
0.8.3
0.8.2
0.8.1
0.8.0
0.7.9
0.7.8
0.7.7
0.7.6
0.7.5
0.7.4
0.7.3
0.7.2
0.7.0
0.6.14
0.6.13
0.6.12
0.6.11
0.6.10
0.6.9
0.6.8
0.6.7
0.6.6
0.6.5
0.6.4
0.6.3
0.6.1
0.6.0
0.5.0
0.4.4
0.4.3
0.4.2
0.4.1
0.3.0
0.2.0
Elixir bindings for llama.cpp — run LLMs locally with Metal, CUDA, Vulkan, or CPU acceleration.
Current section
Files
Jump to
Current section
Files
llama_cpp_ex
Makefile
Makefile
# Makefile for llama_cpp_ex NIF
# Called by elixir_make during `mix compile`
PREFIX = $(MIX_APP_PATH)/priv
BUILD = $(MIX_APP_PATH)/obj
NIF_SO = $(PREFIX)/llama_cpp_ex_nif.so
LLAMA_DIR = $(shell pwd)/vendor/llama.cpp
LLAMA_BUILD = $(BUILD)/llama_build
# Compiler
CXX ?= c++
CXXFLAGS = -std=c++17 -O2 -fPIC -fvisibility=hidden -Wall -Wno-unused-parameter -Wno-unused-function
CXXFLAGS += -I$(ERTS_INCLUDE_DIR)
CXXFLAGS += -I$(FINE_INCLUDE_DIR)
CXXFLAGS += -I$(LLAMA_DIR)/include
CXXFLAGS += -I$(LLAMA_DIR)/ggml/include
CXXFLAGS += -I$(LLAMA_DIR)/common
CXXFLAGS += -I$(LLAMA_DIR)/vendor
# Linker
LDFLAGS = -shared
# Platform detection
UNAME_S := $(shell uname -s)
# Backend selection (auto, metal, cuda, vulkan, cpu)
LLAMA_BACKEND ?= auto
# CMake flags
CMAKE_FLAGS = -DCMAKE_BUILD_TYPE=Release
CMAKE_FLAGS += -DBUILD_SHARED_LIBS=OFF
CMAKE_FLAGS += -DLLAMA_BUILD_EXAMPLES=OFF
CMAKE_FLAGS += -DLLAMA_BUILD_TESTS=OFF
CMAKE_FLAGS += -DLLAMA_BUILD_SERVER=OFF
CMAKE_FLAGS += -DLLAMA_BUILD_TOOLS=OFF
CMAKE_FLAGS += -DLLAMA_OPENSSL=OFF
CMAKE_FLAGS += -DCMAKE_POSITION_INDEPENDENT_CODE=ON
# Backend configuration
ifeq ($(LLAMA_BACKEND),auto)
ifeq ($(UNAME_S),Darwin)
CMAKE_FLAGS += -DGGML_METAL=ON -DGGML_METAL_EMBED_LIBRARY=ON
else
ifneq ($(shell which nvcc 2>/dev/null),)
CMAKE_FLAGS += -DGGML_CUDA=ON
endif
endif
else ifeq ($(LLAMA_BACKEND),metal)
CMAKE_FLAGS += -DGGML_METAL=ON -DGGML_METAL_EMBED_LIBRARY=ON
else ifeq ($(LLAMA_BACKEND),cuda)
CMAKE_FLAGS += -DGGML_CUDA=ON
else ifeq ($(LLAMA_BACKEND),vulkan)
CMAKE_FLAGS += -DGGML_VULKAN=ON
else ifeq ($(LLAMA_BACKEND),cpu)
CMAKE_FLAGS += -DGGML_METAL=OFF -DGGML_CUDA=OFF -DGGML_VULKAN=OFF
endif
# Custom CMake args
ifdef LLAMA_CMAKE_ARGS
CMAKE_FLAGS += $(LLAMA_CMAKE_ARGS)
endif
# Platform-specific linker flags
ifeq ($(UNAME_S),Darwin)
LDFLAGS += -undefined dynamic_lookup
LDFLAGS += -framework Foundation -framework Accelerate
ifneq (,$(filter -DGGML_METAL=ON,$(CMAKE_FLAGS)))
LDFLAGS += -framework Metal -framework MetalKit
endif
else
LDFLAGS += -lstdc++ -lm -lpthread
# ggml-cpu uses OpenMP on Linux when available
ifneq ($(shell $(CXX) -fopenmp -E - < /dev/null 2>/dev/null && echo yes),)
LDFLAGS += -lgomp
endif
endif
# CPU count for parallel builds
NPROC := $(shell nproc 2>/dev/null || sysctl -n hw.ncpu 2>/dev/null || echo 4)
# Sources
NIF_SRC = c_src/llama_cpp_ex/llama_nif.cpp
NIF_OBJ = $(BUILD)/llama_nif.o
# Targets
.PHONY: all clean
all: $(NIF_SO)
# Build llama.cpp static libraries
$(LLAMA_BUILD)/.built: $(LLAMA_DIR)/CMakeLists.txt
@mkdir -p $(LLAMA_BUILD)
cmake -B $(LLAMA_BUILD) -S $(LLAMA_DIR) $(CMAKE_FLAGS)
cmake --build $(LLAMA_BUILD) --config Release -j$(NPROC)
@touch $@
# Compile NIF
$(NIF_OBJ): $(NIF_SRC) c_src/llama_cpp_ex/llama_nif.h $(LLAMA_BUILD)/.built
@mkdir -p $(dir $@)
$(CXX) $(CXXFLAGS) -c $(NIF_SRC) -o $@
# Link NIF - find all static libs from llama.cpp build
$(NIF_SO): $(NIF_OBJ) $(LLAMA_BUILD)/.built
@mkdir -p $(PREFIX)
@LIBS=$$(find $(LLAMA_BUILD) -name '*.a' \
! -path '*/CMakeFiles/*' \
! -path '*/examples/*' \
! -path '*/tests/*' \
| sort); \
if [ "$(UNAME_S)" = "Linux" ]; then \
$(CXX) $(NIF_OBJ) -Wl,--start-group $$LIBS -Wl,--end-group $(LDFLAGS) -o $@; \
else \
$(CXX) $(NIF_OBJ) $$LIBS $(LDFLAGS) -o $@; \
fi
clean:
rm -rf $(BUILD) $(PREFIX)/llama_cpp_ex_nif.so