feat: remove stablediffusion-ggml from main binary (#5861)

* feat: split stablediffusion-ggml from main binary Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Test CI Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Adapt ci tests Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Fixups Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Fixups Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Try to support nvidial4t Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Latest fixups Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
2026-01-07 19:20:04 -06:00 · 2025-07-19 21:58:53 +02:00
parent 1929eb2894
commit 580687da46
14 changed files with 297 additions and 59 deletions
--- a/.github/workflows/backend.yml
+++ b/.github/workflows/backend.yml
@@ -648,6 +648,103 @@ jobs:
            backend: "llama-cpp"
            dockerfile: "./backend/Dockerfile.llama-cpp"
            context: "./"
+          - build-type: 'vulkan'
+            cuda-major-version: ""
+            cuda-minor-version: ""
+            platforms: 'linux/amd64'
+            tag-latest: 'true'
+            tag-suffix: '-gpu-vulkan-llama-cpp'
+            runs-on: 'ubuntu-latest'
+            base-image: "ubuntu:22.04"
+            skip-drivers: 'false'
+            backend: "llama-cpp"
+            dockerfile: "./backend/Dockerfile.llama-cpp"
+            context: "./"
+          # Stablediffusion-ggml
+          - build-type: ''
+            cuda-major-version: ""
+            cuda-minor-version: ""
+            platforms: 'linux/amd64'
+            tag-latest: 'true'
+            tag-suffix: '-cpu-stablediffusion-ggml'
+            runs-on: 'ubuntu-latest'
+            base-image: "ubuntu:22.04"
+            skip-drivers: 'false'
+            backend: "stablediffusion-ggml"
+            dockerfile: "./backend/Dockerfile.go"
+            context: "./"
+          - build-type: 'cublas'
+            cuda-major-version: "12"
+            cuda-minor-version: "0"
+            platforms: 'linux/amd64'
+            tag-latest: 'true'
+            tag-suffix: '-gpu-nvidia-cuda-12-stablediffusion-ggml'
+            runs-on: 'ubuntu-latest'
+            base-image: "ubuntu:22.04"
+            skip-drivers: 'false'
+            backend: "stablediffusion-ggml"
+            dockerfile: "./backend/Dockerfile.go"
+            context: "./"
+          - build-type: 'cublas'
+            cuda-major-version: "11"
+            cuda-minor-version: "7"
+            platforms: 'linux/amd64'
+            tag-latest: 'true'
+            tag-suffix: '-gpu-nvidia-cuda-11-stablediffusion-ggml'
+            runs-on: 'ubuntu-latest'
+            base-image: "ubuntu:22.04"
+            skip-drivers: 'false'
+            backend: "stablediffusion-ggml"
+            dockerfile: "./backend/Dockerfile.go"
+            context: "./"
+          - build-type: 'sycl_f32'
+            cuda-major-version: ""
+            cuda-minor-version: ""
+            platforms: 'linux/amd64'
+            tag-latest: 'true'
+            tag-suffix: '-gpu-intel-sycl-f32-stablediffusion-ggml'
+            runs-on: 'ubuntu-latest'
+            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
+            skip-drivers: 'false'
+            backend: "stablediffusion-ggml"
+            dockerfile: "./backend/Dockerfile.go"
+            context: "./"
+          - build-type: 'sycl_f16'
+            cuda-major-version: ""
+            cuda-minor-version: ""
+            platforms: 'linux/amd64'
+            tag-latest: 'true'
+            tag-suffix: '-gpu-intel-sycl-f16-stablediffusion-ggml'
+            runs-on: 'ubuntu-latest'
+            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
+            skip-drivers: 'false'
+            backend: "stablediffusion-ggml"
+            dockerfile: "./backend/Dockerfile.go"
+            context: "./"
+          - build-type: 'vulkan'
+            cuda-major-version: ""
+            cuda-minor-version: ""
+            platforms: 'linux/amd64'
+            tag-latest: 'true'
+            tag-suffix: '-gpu-vulkan-stablediffusion-ggml'
+            runs-on: 'ubuntu-latest'
+            base-image: "ubuntu:22.04"
+            skip-drivers: 'false'
+            backend: "stablediffusion-ggml"
+            dockerfile: "./backend/Dockerfile.go"
+            context: "./"
+          - build-type: 'cublas'
+            cuda-major-version: "12"
+            cuda-minor-version: "0"
+            platforms: 'linux/arm64'
+            skip-drivers: 'true'
+            tag-latest: 'auto'
+            tag-suffix: '-nvidia-l4t-arm64-stablediffusion-ggml'
+            base-image: "nvcr.io/nvidia/l4t-jetpack:r36.4.0"
+            runs-on: 'ubuntu-24.04-arm'
+            backend: "stablediffusion-ggml"
+            dockerfile: "./backend/Dockerfile.go"
+            context: "./"
  llama-cpp-darwin:
    runs-on: macOS-14
    strategy:
--- a/.github/workflows/backend_build.yml
+++ b/.github/workflows/backend_build.yml
@@ -64,7 +64,7 @@ on:
        required: true

 jobs:
-  reusable_python_backend-build:
+  backend-build:
    runs-on: ${{ inputs.runs-on }}
    steps:

--- a/.github/workflows/bump_deps.yaml
+++ b/.github/workflows/bump_deps.yaml
@@ -21,14 +21,10 @@ jobs:
            variable: "BARKCPP_VERSION"
            branch: "main"
            file: "Makefile"
-          - repository: "leejet/stable-diffusion.cpp"
+          - repository: "richiejp/stable-diffusion.cpp"
            variable: "STABLEDIFFUSION_GGML_VERSION"
            branch: "master"
-            file: "Makefile"
-          - repository: "mudler/go-stable-diffusion"
-            variable: "STABLEDIFFUSION_VERSION"
-            branch: "master"
-            file: "Makefile"
+            file: "backend/go/stablediffusion-ggml/Makefile"
          - repository: "mudler/go-piper"
            variable: "PIPER_VERSION"
            branch: "master"
--- a/.github/workflows/test.yml
+++ b/.github/workflows/test.yml
@@ -103,7 +103,7 @@ jobs:

          make -C backend/python/transformers

-          make backends/llama-cpp backends/piper
+          make backends/llama-cpp backends/piper backends/stablediffusion-ggml
        env:
          CUDA_VERSION: 12-4
      - name: Test
@@ -168,7 +168,7 @@ jobs:
          PATH="$PATH:$HOME/go/bin" make protogen-go
      - name: Test
        run: |
-            PATH="$PATH:$HOME/go/bin" make backends/llama-cpp backends/piper docker-build-aio e2e-aio
+            PATH="$PATH:$HOME/go/bin" make backends/llama-cpp backends/piper backends/stablediffusion-ggml docker-build-aio e2e-aio
      - name: Setup tmate session if tests fail
        if: ${{ failure() }}
        uses: mxschmitt/action-tmate@v3.22
--- a/37
+++ b/37
@@ -9,10 +9,6 @@ DETECT_LIBS?=true
 WHISPER_REPO?=https://github.com/ggml-org/whisper.cpp
 WHISPER_CPP_VERSION?=032697b9a850dc2615555e2a93a683cc3dd58559

-# stablediffusion.cpp (ggml)
-STABLEDIFFUSION_GGML_REPO?=https://github.com/richiejp/stable-diffusion.cpp
-STABLEDIFFUSION_GGML_VERSION?=53e3b17eb3d0b5760ced06a1f98320b68b34aaae
-
 # ONEAPI variables for SYCL
 export ONEAPI_VARS?=/opt/intel/oneapi/setvars.sh
 ONEAPI_VERSION=2025.1
@@ -207,13 +203,6 @@ endif

 ALL_GRPC_BACKENDS=backend-assets/grpc/huggingface
 ALL_GRPC_BACKENDS+=backend-assets/grpc/whisper
-
-ifeq ($(ONNX_OS),linux)
-ifeq ($(ONNX_ARCH),x64)
-	ALL_GRPC_BACKENDS+=backend-assets/grpc/stablediffusion-ggml
-endif
-endif
-
 ALL_GRPC_BACKENDS+=backend-assets/grpc/local-store
 ALL_GRPC_BACKENDS+=backend-assets/grpc/silero-vad
 ALL_GRPC_BACKENDS+=$(OPTIONAL_GRPC)
@@ -236,20 +225,6 @@ endif

 all: help

-## stablediffusion (ggml)
-sources/stablediffusion-ggml.cpp:
-	git clone --recursive $(STABLEDIFFUSION_GGML_REPO) sources/stablediffusion-ggml.cpp && \
-	cd sources/stablediffusion-ggml.cpp && \
-	git checkout $(STABLEDIFFUSION_GGML_VERSION) && \
-	git submodule update --init --recursive --depth 1 --single-branch
-
-backend/go/image/stablediffusion-ggml/libsd.a: sources/stablediffusion-ggml.cpp
-	$(MAKE) -C backend/go/image/stablediffusion-ggml build/libstable-diffusion.a
-	$(MAKE) -C backend/go/image/stablediffusion-ggml libsd.a
-
-backend-assets/grpc/stablediffusion-ggml: backend/go/image/stablediffusion-ggml/libsd.a backend-assets/grpc
-	$(MAKE) -C backend/go/image/stablediffusion-ggml CGO_LDFLAGS="$(CGO_LDFLAGS)" stablediffusion-ggml
-
 sources/onnxruntime:
 	mkdir -p sources/onnxruntime
 	curl -L https://github.com/microsoft/onnxruntime/releases/download/v$(ONNX_VERSION)/onnxruntime-$(ONNX_OS)-$(ONNX_ARCH)-$(ONNX_VERSION).tgz -o sources/onnxruntime/onnxruntime-$(ONNX_OS)-$(ONNX_ARCH)-$(ONNX_VERSION).tgz
@@ -278,7 +253,7 @@ sources/whisper.cpp/build/src/libwhisper.a: sources/whisper.cpp
 	cd sources/whisper.cpp && cmake $(WHISPER_CMAKE_ARGS) . -B ./build
 	cd sources/whisper.cpp/build && cmake --build . --config Release

-get-sources: sources/stablediffusion-ggml.cpp sources/whisper.cpp
+get-sources: sources/whisper.cpp

 replace:
 	$(GOCMD) mod edit -replace github.com/ggerganov/whisper.cpp=$(CURDIR)/sources/whisper.cpp
@@ -307,7 +282,6 @@ clean: ## Remove build related file
 	rm -rf release/
 	rm -rf backend-assets/*
 	$(MAKE) -C backend/cpp/grpc clean
-	$(MAKE) -C backend/go/image/stablediffusion-ggml clean
 	$(MAKE) dropreplace
 	$(MAKE) protogen-clean
 	rmdir pkg/grpc/proto || true
@@ -401,6 +375,9 @@ backends/llama-cpp: docker-build-llama-cpp docker-save-llama-cpp build-api
 backends/piper: docker-build-piper docker-save-piper build-api
 	./local-ai backends install "ocifile://$(abspath ./backend-images/piper.tar)"

+backends/stablediffusion-ggml: docker-build-stablediffusion-ggml docker-save-stablediffusion-ggml build-api
+	./local-ai backends install "ocifile://$(abspath ./backend-images/stablediffusion-ggml.tar)"
+
 ########################################################
 ## AIO tests
 ########################################################
@@ -736,6 +713,12 @@ docker-save-llama-cpp: backend-images
 docker-save-bark-cpp: backend-images
 	docker save local-ai-backend:bark-cpp -o backend-images/bark-cpp.tar

+docker-build-stablediffusion-ggml:
+	docker build -t local-ai-backend:stablediffusion-ggml -f backend/Dockerfile.go --build-arg BACKEND=stablediffusion-ggml .
+
+docker-save-stablediffusion-ggml: backend-images
+	docker save local-ai-backend:stablediffusion-ggml -o backend-images/stablediffusion-ggml.tar
+
 docker-build-rerankers:
 	docker build -t local-ai-backend:rerankers -f backend/Dockerfile.python --build-arg BACKEND=rerankers .

--- a/backend/cpp/llama-cpp/Makefile
+++ b/backend/cpp/llama-cpp/Makefile
@@ -35,6 +35,8 @@ else ifeq ($(BUILD_TYPE),hipblas)
 #	AMDGPU_TARGETS ?= "$(GPU_TARGETS)"
 	CMAKE_ARGS+=-DGGML_HIP=ON
 #	CMAKE_ARGS+=-DGGML_HIP=ON -DAMDGPU_TARGETS="$(AMDGPU_TARGETS)" -DGPU_TARGETS="$(GPU_TARGETS)"
+else ifeq ($(BUILD_TYPE),vulkan)
+	CMAKE_ARGS+=-DGGML_VULKAN=1
 else ifeq ($(OS),Darwin)
 	ifeq ($(BUILD_TYPE),)
 		BUILD_TYPE=metal
--- a/backend/go/image/stablediffusion-ggml/Makefile
+++ b/backend/go/image/stablediffusion-ggml/Makefile
@@ -4,9 +4,11 @@ LIBRARY_PATH := $(abspath ./)
 AR?=ar
 CMAKE_ARGS?=
 BUILD_TYPE?=
+NATIVE?=false
+CUDA_LIBPATH?=/usr/local/cuda/lib64/
 ONEAPI_VARS?=/opt/intel/oneapi/setvars.sh
 # keep standard at C11 and C++11
-CXXFLAGS = -I. -I$(INCLUDE_PATH)/../../../../sources/stablediffusion-ggml.cpp/thirdparty -I$(INCLUDE_PATH)/../../../../sources/stablediffusion-ggml.cpp/ggml/include -I$(INCLUDE_PATH)/../../../../sources/stablediffusion-ggml.cpp -O3 -DNDEBUG -std=c++17 -fPIC
+CXXFLAGS = -I. -I$(INCLUDE_PATH)/sources/stablediffusion-ggml.cpp/thirdparty -I$(INCLUDE_PATH)/sources/stablediffusion-ggml.cpp/ggml/include -I$(INCLUDE_PATH)/sources/stablediffusion-ggml.cpp -O3 -DNDEBUG -std=c++17 -fPIC

 GOCMD?=go
 CGO_LDFLAGS?=
@@ -15,12 +17,21 @@ CGO_LDFLAGS_SYCL=
 GO_TAGS?=
 LD_FLAGS?=

+# stablediffusion.cpp (ggml)
+STABLEDIFFUSION_GGML_REPO?=https://github.com/richiejp/stable-diffusion.cpp
+STABLEDIFFUSION_GGML_VERSION?=53e3b17eb3d0b5760ced06a1f98320b68b34aaae
+
 # Disable Shared libs as we are linking on static gRPC and we can't mix shared and static
 CMAKE_ARGS+=-DBUILD_SHARED_LIBS=OFF

+ifeq ($(NATIVE),false)
+	CMAKE_ARGS+=-DGGML_NATIVE=OFF
+endif
+
 # If build type is cublas, then we set -DGGML_CUDA=ON to CMAKE_ARGS automatically
 ifeq ($(BUILD_TYPE),cublas)
-	CMAKE_ARGS+=-DSD_CUDA=ON
+	CMAKE_ARGS+=-DSD_CUDA=ON -DGGML_CUDA=ON
+	CGO_LDFLAGS+=-lcublas -lcudart -L$(CUDA_LIBPATH) -L$(CUDA_LIBPATH)/stubs/ -lcuda
 # If build type is openblas then we set -DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS
 # to CMAKE_ARGS automatically
 else ifeq ($(BUILD_TYPE),openblas)
@@ -30,14 +41,17 @@ else ifeq ($(BUILD_TYPE),clblas)
 	CMAKE_ARGS+=-DGGML_CLBLAST=ON -DCLBlast_DIR=/some/path
 # If it's hipblas we do have also to set CC=/opt/rocm/llvm/bin/clang CXX=/opt/rocm/llvm/bin/clang++
 else ifeq ($(BUILD_TYPE),hipblas)
-	CMAKE_ARGS+=-DSD_HIPBLAS=ON
+	CMAKE_ARGS+=-DSD_HIPBLAS=ON -DGGML_HIPBLAS=ON
 # If it's OSX, DO NOT embed the metal library - -DGGML_METAL_EMBED_LIBRARY=ON requires further investigation
 # But if it's OSX without metal, disable it here
+else ifeq ($(BUILD_TYPE),vulkan)
+	CMAKE_ARGS+=-DSD_VULKAN=ON -DGGML_VULKAN=ON
+	CGO_LDFLAGS+=-lvulkan
 else ifeq ($(OS),Darwin)
 	ifneq ($(BUILD_TYPE),metal)
-		CMAKE_ARGS+=-DSD_METAL=OFF
+		CMAKE_ARGS+=-DSD_METAL=OFF -DGGML_METAL=OFF
 	else
-		CMAKE_ARGS+=-DSD_METAL=ON
+		CMAKE_ARGS+=-DSD_METAL=ON -DGGML_METAL=ON
 		CMAKE_ARGS+=-DGGML_METAL_EMBED_LIBRARY=ON
 		TARGET+=--target ggml-metal
 	endif
@@ -49,8 +63,8 @@ ifeq ($(BUILD_TYPE),sycl_f16)
 		-DCMAKE_CXX_COMPILER=icpx \
 		-DSD_SYCL=ON \
 		-DGGML_SYCL_F16=ON
-	CC=icx
-	CXX=icpx
+	export CC=icx
+	export CXX=icpx
 	CGO_LDFLAGS_SYCL += -fsycl -L${DNNLROOT}/lib -ldnnl ${MKLROOT}/lib/intel64/libmkl_sycl.a -fiopenmp -fopenmp-targets=spir64 -lOpenCL
 	CGO_LDFLAGS_SYCL += $(shell pkg-config --libs mkl-static-lp64-gomp)
 	CGO_CXXFLAGS += -fiopenmp -fopenmp-targets=spir64
@@ -62,8 +76,8 @@ ifeq ($(BUILD_TYPE),sycl_f32)
 		-DCMAKE_C_COMPILER=icx \
 		-DCMAKE_CXX_COMPILER=icpx \
 		-DSD_SYCL=ON
-	CC=icx
-	CXX=icpx
+	export CC=icx
+	export CXX=icpx
 	CGO_LDFLAGS_SYCL += -fsycl -L${DNNLROOT}/lib -ldnnl ${MKLROOT}/lib/intel64/libmkl_sycl.a -fiopenmp -fopenmp-targets=spir64 -lOpenCL
 	CGO_LDFLAGS_SYCL += $(shell pkg-config --libs mkl-static-lp64-gomp)
 	CGO_CXXFLAGS += -fiopenmp -fopenmp-targets=spir64
@@ -101,12 +115,12 @@ ifneq (,$(findstring sycl,$(BUILD_TYPE)))
 	+bash -c "source $(ONEAPI_VARS); \
 	mkdir -p build && \
 	cd build && \
-	cmake $(CMAKE_ARGS) ../../../../../sources/stablediffusion-ggml.cpp && \
+	cmake $(CMAKE_ARGS) ../sources/stablediffusion-ggml.cpp && \
 	cmake --build . --config Release"
 else
 	mkdir -p build && \
 	cd build && \
-	cmake $(CMAKE_ARGS) ../../../../../sources/stablediffusion-ggml.cpp && \
+	cmake $(CMAKE_ARGS) ../sources/stablediffusion-ggml.cpp && \
 	cmake --build . --config Release
 endif
 	$(MAKE) $(COMBINED_LIB)
@@ -119,17 +133,26 @@ else
 	$(CXX) $(CXXFLAGS) gosd.cpp -o gosd.o -c
 endif

-libsd.a: gosd.o
+## stablediffusion (ggml)
+sources/stablediffusion-ggml.cpp:
+	git clone --recursive $(STABLEDIFFUSION_GGML_REPO) sources/stablediffusion-ggml.cpp && \
+	cd sources/stablediffusion-ggml.cpp && \
+	git checkout $(STABLEDIFFUSION_GGML_VERSION) && \
+	git submodule update --init --recursive --depth 1 --single-branch
+
+libsd.a: sources/stablediffusion-ggml.cpp build/libstable-diffusion.a gosd.o
 	cp $(INCLUDE_PATH)/build/libstable-diffusion.a ./libsd.a
 	$(AR) rcs libsd.a gosd.o

-stablediffusion-ggml:
+stablediffusion-ggml: libsd.a
 	CGO_LDFLAGS="$(CGO_LDFLAGS) $(CGO_LDFLAGS_SYCL)" C_INCLUDE_PATH="$(INCLUDE_PATH)" LIBRARY_PATH="$(LIBRARY_PATH)" \
 	CC="$(CC)" CXX="$(CXX)" CGO_CXXFLAGS="$(CGO_CXXFLAGS)" \
-	$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o ../../../../backend-assets/grpc/stablediffusion-ggml ./
-ifneq ($(UPX),)
-	$(UPX) ../../../../backend-assets/grpc/stablediffusion-ggml
-endif
+	$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o stablediffusion-ggml ./
+
+package:
+	bash package.sh
+
+build: stablediffusion-ggml package

 clean:
 	rm -rf gosd.o libsd.a build $(COMBINED_LIB)
--- a/backend/go/image/stablediffusion-ggml/gosd.cpp
+++ b/backend/go/image/stablediffusion-ggml/gosd.cpp
--- a/backend/go/image/stablediffusion-ggml/gosd.go
+++ b/backend/go/image/stablediffusion-ggml/gosd.go
@@ -1,6 +1,6 @@
 package main

-// #cgo CXXFLAGS: -I${SRCDIR}/../../../../sources/stablediffusion-ggml.cpp/thirdparty -I${SRCDIR}/../../../../sources/stablediffusion-ggml.cpp -I${SRCDIR}/../../../../sources/stablediffusion-ggml.cpp/ggml/include
+// #cgo CXXFLAGS: -I${SRCDIR}/sources/stablediffusion-ggml.cpp/thirdparty -I${SRCDIR}/sources/stablediffusion-ggml.cpp -I${SRCDIR}/sources/stablediffusion-ggml.cpp/ggml/include
 // #cgo LDFLAGS: -L${SRCDIR}/ -lsd -lstdc++ -lm -lggmlall -lgomp
 // #include <gosd.h>
 // #include <stdlib.h>
--- a/backend/go/image/stablediffusion-ggml/gosd.h
+++ b/backend/go/image/stablediffusion-ggml/gosd.h
--- a/backend/go/image/stablediffusion-ggml/main.go
+++ b/backend/go/image/stablediffusion-ggml/main.go
--- a/backend/go/stablediffusion-ggml/package.sh
+++ b/backend/go/stablediffusion-ggml/package.sh
@@ -0,0 +1,52 @@
+#!/bin/bash
+
+# Script to copy the appropriate libraries based on architecture
+# This script is used in the final stage of the Dockerfile
+
+set -e
+
+CURDIR=$(dirname "$(realpath $0)")
+
+# Create lib directory
+mkdir -p $CURDIR/package/lib
+
+cp -avrf $CURDIR/stablediffusion-ggml $CURDIR/package/
+cp -rfv $CURDIR/run.sh $CURDIR/package/
+
+# Detect architecture and copy appropriate libraries
+if [ -f "/lib64/ld-linux-x86-64.so.2" ]; then
+    # x86_64 architecture
+    echo "Detected x86_64 architecture, copying x86_64 libraries..."
+    cp -arfLv /lib64/ld-linux-x86-64.so.2 $CURDIR/package/lib/ld.so
+    cp -arfLv /lib/x86_64-linux-gnu/libc.so.6 $CURDIR/package/lib/libc.so.6
+    cp -arfLv /lib/x86_64-linux-gnu/libgcc_s.so.1 $CURDIR/package/lib/libgcc_s.so.1
+    cp -arfLv /lib/x86_64-linux-gnu/libstdc++.so.6 $CURDIR/package/lib/libstdc++.so.6
+    cp -arfLv /lib/x86_64-linux-gnu/libm.so.6 $CURDIR/package/lib/libm.so.6
+    cp -arfLv /lib/x86_64-linux-gnu/libgomp.so.1 $CURDIR/package/lib/libgomp.so.1
+    cp -arfLv /lib/x86_64-linux-gnu/libgcc_s.so.1 $CURDIR/package/lib/libgcc_s.so.1
+    cp -arfLv /lib/x86_64-linux-gnu/libstdc++.so.6 $CURDIR/package/lib/libstdc++.so.6
+    cp -arfLv /lib/x86_64-linux-gnu/libdl.so.2 $CURDIR/package/lib/libdl.so.2
+    cp -arfLv /lib/x86_64-linux-gnu/librt.so.1 $CURDIR/package/lib/librt.so.1
+    cp -arfLv /lib/x86_64-linux-gnu/libpthread.so.0 $CURDIR/package/lib/libpthread.so.0
+elif [ -f "/lib/ld-linux-aarch64.so.1" ]; then
+    # ARM64 architecture
+    echo "Detected ARM64 architecture, copying ARM64 libraries..."
+    cp -arfLv /lib/ld-linux-aarch64.so.1 $CURDIR/package/lib/ld.so
+    cp -arfLv /lib/aarch64-linux-gnu/libc.so.6 $CURDIR/package/lib/libc.so.6
+    cp -arfLv /lib/aarch64-linux-gnu/libgcc_s.so.1 $CURDIR/package/lib/libgcc_s.so.1
+    cp -arfLv /lib/aarch64-linux-gnu/libstdc++.so.6 $CURDIR/package/lib/libstdc++.so.6
+    cp -arfLv /lib/aarch64-linux-gnu/libm.so.6 $CURDIR/package/lib/libm.so.6
+    cp -arfLv /lib/aarch64-linux-gnu/libgomp.so.1 $CURDIR/package/lib/libgomp.so.1
+    cp -arfLv /lib/aarch64-linux-gnu/libgcc_s.so.1 $CURDIR/package/lib/libgcc_s.so.1
+    cp -arfLv /lib/aarch64-linux-gnu/libstdc++.so.6 $CURDIR/package/lib/libstdc++.so.6
+    cp -arfLv /lib/aarch64-linux-gnu/libdl.so.2 $CURDIR/package/lib/libdl.so.2
+    cp -arfLv /lib/aarch64-linux-gnu/librt.so.1 $CURDIR/package/lib/librt.so.1
+    cp -arfLv /lib/aarch64-linux-gnu/libpthread.so.0 $CURDIR/package/lib/libpthread.so.0
+else
+    echo "Error: Could not detect architecture"
+    exit 1
+fi
+
+echo "Packaging completed successfully" 
+ls -liah $CURDIR/package/
+ls -liah $CURDIR/package/lib/
--- a/backend/go/stablediffusion-ggml/run.sh
+++ b/backend/go/stablediffusion-ggml/run.sh
@@ -0,0 +1,14 @@
+#!/bin/bash
+set -ex
+
+CURDIR=$(dirname "$(realpath $0)")
+
+export LD_LIBRARY_PATH=$CURDIR/lib:$LD_LIBRARY_PATH
+
+# If there is a lib/ld.so, use it
+if [ -f $CURDIR/lib/ld.so ]; then
+	echo "Using lib/ld.so"
+	exec $CURDIR/lib/ld.so $CURDIR/stablediffusion-ggml "$@"
+fi
+
+exec $CURDIR/stablediffusion-ggml "$@"
--- a/backend/index.yaml
+++ b/backend/index.yaml
@@ -23,8 +23,34 @@
    intel: "intel-sycl-f16-llama-cpp"
    amd: "rocm-llama-cpp"
    metal: "metal-llama-cpp"
+    vulkan: "vulkan-llama-cpp"
    nvidia-l4t: "nvidia-l4t-arm64-llama-cpp"
    darwin-x86: "darwin-x86-llama-cpp"
+- &stablediffusionggml
+  name: "stablediffusion-ggml"
+  alias: "stablediffusion-ggml"
+  license: mit
+  icon: https://github.com/leejet/stable-diffusion.cpp/raw/master/assets/cat_with_sd_cpp_42.png
+  description: |
+     Stable Diffusion and Flux in pure C/C++
+  urls:
+    - https://github.com/leejet/stable-diffusion.cpp
+  tags:
+    - image-generation
+    - CPU
+    - GPU
+    - Metal
+    - CUDA
+    - HIP
+  capabilities:
+    default: "cpu-stablediffusion-ggml"
+    nvidia: "cuda12-stablediffusion-ggml"
+    intel: "intel-sycl-f16-stablediffusion-ggml"
+    #amd: "rocm-stablediffusion-ggml"
+    vulkan: "vulkan-stablediffusion-ggml"
+    nvidia-l4t: "nvidia-l4t-arm64-stablediffusion-ggml"
+    # metal: "metal-stablediffusion-ggml"
+    # darwin-x86: "darwin-x86-stablediffusion-ggml"
 - &vllm
  name: "vllm"
  license: apache-2.0
@@ -237,10 +263,6 @@
  tags:
    - text-to-speech
    - TTS
-  capabilities:
-    nvidia: "cuda12-piper"
-    intel: "intel-sycl-f32-piper"
-    amd: "rocm-piper"
 - !!merge <<: *piper
  name: "piper-development"
  uri: "quay.io/go-skynet/local-ai-backends:master-piper"
@@ -278,6 +300,12 @@
 - !!merge <<: *llamacpp
  name: "intel-sycl-f16-llama-cpp"
  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f16-llama-cpp"
+- !!merge <<: *llamacpp
+  name: "vulkan-llama-cpp"
+  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-vulkan-llama-cpp"
+- !!merge <<: *llamacpp
+  name: "vulkan-llama-cpp-development"
+  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-vulkan-llama-cpp"
 - !!merge <<: *llamacpp
  name: "metal-llama-cpp"
  uri: "quay.io/go-skynet/local-ai-backends:latest-metal-darwin-arm64-llama-cpp"
@@ -297,8 +325,51 @@
  name: "intel-sycl-f32-llama-cpp-development"
  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f32-llama-cpp"
 - !!merge <<: *llamacpp
-  name: "intel-sycl-f16-vllm-development"
+  name: "intel-sycl-f16-llama-cpp-development"
  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f16-vllm"
+## stablediffusion-ggml
+- !!merge <<: *stablediffusionggml
+  name: "cpu-stablediffusion-ggml"
+  uri: "quay.io/go-skynet/local-ai-backends:latest-cpu-stablediffusion-ggml"
+- !!merge <<: *stablediffusionggml
+  name: "cpu-stablediffusion-ggml-development"
+  uri: "quay.io/go-skynet/local-ai-backends:master-cpu-stablediffusion-ggml"
+- !!merge <<: *stablediffusionggml
+  name: "vulkan-stablediffusion-ggml"
+  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-vulkan-stablediffusion-ggml"
+- !!merge <<: *stablediffusionggml
+  name: "vulkan-stablediffusion-ggml-development"
+  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-vulkan-stablediffusion-ggml"
+- !!merge <<: *stablediffusionggml
+  name: "cuda12-stablediffusion-ggml"
+  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-nvidia-cuda-12-stablediffusion-ggml"
+- !!merge <<: *stablediffusionggml
+  name: "intel-sycl-f32-stablediffusion-ggml"
+  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f32-stablediffusion-ggml"
+- !!merge <<: *stablediffusionggml
+  name: "intel-sycl-f16-stablediffusion-ggml"
+  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f16-stablediffusion-ggml"
+- !!merge <<: *stablediffusionggml
+  name: "cuda11-stablediffusion-ggml"
+  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-nvidia-cuda-11-stablediffusion-ggml"
+- !!merge <<: *stablediffusionggml
+  name: "cuda12-stablediffusion-ggml-development"
+  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-12-stablediffusion-ggml"
+- !!merge <<: *stablediffusionggml
+  name: "intel-sycl-f32-stablediffusion-ggml-development"
+  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f32-stablediffusion-ggml"
+- !!merge <<: *stablediffusionggml
+  name: "intel-sycl-f16-stablediffusion-ggml-development"
+  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f16-stablediffusion-ggml"
+- !!merge <<: *stablediffusionggml
+  name: "cuda11-stablediffusion-ggml-development"
+  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-11-stablediffusion-ggml"
+- !!merge <<: *stablediffusionggml
+  name: "nvidia-l4t-arm64-stablediffusion-ggml-development"
+  uri: "quay.io/go-skynet/local-ai-backends:master-nvidia-l4t-arm64-stablediffusion-ggml"
+- !!merge <<: *stablediffusionggml
+  name: "nvidia-l4t-arm64-stablediffusion-ggml"
+  uri: "quay.io/go-skynet/local-ai-backends:latest-nvidia-l4t-arm64-stablediffusion-ggml"
 # vllm
 - !!merge <<: *vllm
  name: "vllm-development"