diff --git a/.devops/main-intel.Dockerfile b/.devops/main-intel.Dockerfile
index 572e5d8ea..274b91b71 100644
--- a/.devops/main-intel.Dockerfile
+++ b/.devops/main-intel.Dockerfile
@@ -10,14 +10,12 @@ WORKDIR /app
 
 COPY . .
 
-RUN mkdir build && \
-    cd build && \
-    if [ "${LLAMA_SYCL_F16}" = "ON" ]; then \
+RUN if [ "${LLAMA_SYCL_F16}" = "ON" ]; then \
         echo "LLAMA_SYCL_F16 is set" && \
         export OPT_SYCL_F16="-DLLAMA_SYCL_F16=ON"; \
     fi && \
-    cmake .. -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx ${OPT_SYCL_F16} && \
-    cmake --build . --config Release --target main
+    cmake -B build -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx ${OPT_SYCL_F16} && \
+    cmake --build build --config Release --target main
 
 FROM intel/oneapi-basekit:$ONEAPI_VERSION as runtime
 
diff --git a/.devops/main-vulkan.Dockerfile b/.devops/main-vulkan.Dockerfile
index bca460365..6c2b2ed5b 100644
--- a/.devops/main-vulkan.Dockerfile
+++ b/.devops/main-vulkan.Dockerfile
@@ -14,10 +14,8 @@ RUN wget -qO - https://packages.lunarg.com/lunarg-signing-key-pub.asc | apt-key
 # Build it
 WORKDIR /app
 COPY . .
-RUN mkdir build && \
-    cd build && \
-    cmake .. -DLLAMA_VULKAN=1 && \
-    cmake --build . --config Release --target main
+RUN cmake -B build -DLLAMA_VULKAN=1 && \
+    cmake --build build --config Release --target main
 
 # Clean up
 WORKDIR /
diff --git a/.devops/server-intel.Dockerfile b/.devops/server-intel.Dockerfile
index 304487335..a8e451fa9 100644
--- a/.devops/server-intel.Dockerfile
+++ b/.devops/server-intel.Dockerfile
@@ -10,14 +10,12 @@ WORKDIR /app
 
 COPY . .
 
-RUN mkdir build && \
-    cd build && \
-    if [ "${LLAMA_SYCL_F16}" = "ON" ]; then \
+RUN if [ "${LLAMA_SYCL_F16}" = "ON" ]; then \
         echo "LLAMA_SYCL_F16 is set" && \
         export OPT_SYCL_F16="-DLLAMA_SYCL_F16=ON"; \
     fi && \
-    cmake .. -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_CURL=ON ${OPT_SYCL_F16} && \
-    cmake --build . --config Release --target server
+    cmake -B build -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_CURL=ON ${OPT_SYCL_F16} && \
+    cmake --build build --config Release --target server
 
 FROM intel/oneapi-basekit:$ONEAPI_VERSION as runtime
 
diff --git a/.devops/server-vulkan.Dockerfile b/.devops/server-vulkan.Dockerfile
index 7e5a5283b..6e757e171 100644
--- a/.devops/server-vulkan.Dockerfile
+++ b/.devops/server-vulkan.Dockerfile
@@ -18,10 +18,8 @@ RUN apt-get update && \
 # Build it
 WORKDIR /app
 COPY . .
-RUN mkdir build && \
-    cd build && \
-    cmake .. -DLLAMA_VULKAN=1 -DLLAMA_CURL=1 && \
-    cmake --build . --config Release --target server
+RUN cmake -B build -DLLAMA_VULKAN=1 -DLLAMA_CURL=1 && \
+    cmake --build build --config Release --target server
 
 # Clean up
 WORKDIR /
diff --git a/.github/workflows/bench.yml b/.github/workflows/bench.yml
index 6dc91d27b..3e968d179 100644
--- a/.github/workflows/bench.yml
+++ b/.github/workflows/bench.yml
@@ -96,9 +96,7 @@ jobs:
         id: cmake_build
         run: |
           set -eux
-          mkdir build
-          cd build
-          cmake .. \
+          cmake -B build \
               -DLLAMA_NATIVE=OFF \
               -DLLAMA_BUILD_SERVER=ON \
               -DLLAMA_CURL=ON \
@@ -109,7 +107,7 @@ jobs:
               -DLLAMA_FATAL_WARNINGS=OFF \
               -DLLAMA_ALL_WARNINGS=OFF \
               -DCMAKE_BUILD_TYPE=Release;
-          cmake --build . --config Release -j $(nproc) --target server
+          cmake --build build --config Release -j $(nproc) --target server
 
       - name: Download the dataset
         id: download_dataset
diff --git a/.github/workflows/server.yml b/.github/workflows/server.yml
index 79cd7d643..afac89c5b 100644
--- a/.github/workflows/server.yml
+++ b/.github/workflows/server.yml
@@ -94,15 +94,13 @@ jobs:
       - name: Build
         id: cmake_build
         run: |
-          mkdir build
-          cd build
-          cmake .. \
+          cmake -B build \
               -DLLAMA_NATIVE=OFF \
               -DLLAMA_BUILD_SERVER=ON \
               -DLLAMA_CURL=ON \
               -DCMAKE_BUILD_TYPE=${{ matrix.build_type }} \
               -DLLAMA_SANITIZE_${{ matrix.sanitizer }}=ON ;
-          cmake --build . --config ${{ matrix.build_type }} -j $(nproc) --target server
+          cmake --build build --config ${{ matrix.build_type }} -j $(nproc) --target server
 
 
       - name: Tests
@@ -143,10 +141,8 @@ jobs:
       - name: Build
         id: cmake_build
         run: |
-          mkdir build
-          cd build
-          cmake .. -DLLAMA_CURL=ON -DCURL_LIBRARY="$env:RUNNER_TEMP/libcurl/lib/libcurl.dll.a" -DCURL_INCLUDE_DIR="$env:RUNNER_TEMP/libcurl/include"
-          cmake --build . --config Release -j ${env:NUMBER_OF_PROCESSORS} --target server
+          cmake -B build -DLLAMA_CURL=ON -DCURL_LIBRARY="$env:RUNNER_TEMP/libcurl/lib/libcurl.dll.a" -DCURL_INCLUDE_DIR="$env:RUNNER_TEMP/libcurl/include"
+          cmake --build build --config Release -j ${env:NUMBER_OF_PROCESSORS} --target server
 
       - name: Python setup
         id: setup_python
diff --git a/README-sycl.md b/README-sycl.md
index dc98c7b3e..cfa248a95 100644
--- a/README-sycl.md
+++ b/README-sycl.md
@@ -185,9 +185,8 @@ Upon a successful installation, SYCL is enabled for the available intel devices,
 ```sh
 git clone https://github.com/oneapi-src/oneMKL
 cd oneMKL
-mkdir -p buildWithCublas && cd buildWithCublas
-cmake ../ -DCMAKE_CXX_COMPILER=icpx -DCMAKE_C_COMPILER=icx -DENABLE_MKLGPU_BACKEND=OFF -DENABLE_MKLCPU_BACKEND=OFF -DENABLE_CUBLAS_BACKEND=ON -DTARGET_DOMAINS=blas
-make
+cmake -B buildWithCublas -DCMAKE_CXX_COMPILER=icpx -DCMAKE_C_COMPILER=icx -DENABLE_MKLGPU_BACKEND=OFF -DENABLE_MKLCPU_BACKEND=OFF -DENABLE_CUBLAS_BACKEND=ON -DTARGET_DOMAINS=blas
+cmake --build buildWithCublas --config Release
 ```
 
 
@@ -227,16 +226,15 @@ Similarly, user targeting Nvidia GPUs should expect at least one SYCL-CUDA devic
 source /opt/intel/oneapi/setvars.sh
 
 # Build LLAMA with MKL BLAS acceleration for intel GPU
-mkdir -p build && cd build
 
 # Option 1: Use FP32 (recommended for better performance in most cases)
-cmake .. -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx
+cmake -B build -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx
 
 # Option 2: Use FP16
-cmake .. -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_SYCL_F16=ON
+cmake -B build -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_SYCL_F16=ON
 
-#build all binary
-cmake --build . --config Release -j -v
+# build all binary
+cmake --build build --config Release -j -v
 ```
 
 #### Nvidia GPU
@@ -248,16 +246,15 @@ export CPLUS_INCLUDE_DIR=/path/to/oneMKL/buildWithCublas/include:$CPLUS_INCLUDE_
 export CPLUS_INCLUDE_DIR=/path/to/oneMKL/include:$CPLUS_INCLUDE_DIR
 
 # Build LLAMA with Nvidia BLAS acceleration through SYCL
-mkdir -p build && cd build
 
 # Option 1: Use FP32 (recommended for better performance in most cases)
-cmake .. -DLLAMA_SYCL=ON -DLLAMA_SYCL_TARGET=NVIDIA -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx
+cmake -B build -DLLAMA_SYCL=ON -DLLAMA_SYCL_TARGET=NVIDIA -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx
 
 # Option 2: Use FP16
-cmake .. -DLLAMA_SYCL=ON -DLLAMA_SYCL_TARGET=NVIDIA -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_SYCL_F16=ON
+cmake -B build -DLLAMA_SYCL=ON -DLLAMA_SYCL_TARGET=NVIDIA -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_SYCL_F16=ON
 
-#build all binary
-cmake --build . --config Release -j -v
+# build all binary
+cmake --build build --config Release -j -v
 
 ```
 
@@ -412,17 +409,15 @@ b. Download & install mingw-w64 make for Windows provided by w64devkit
 On the oneAPI command line window, step into the llama.cpp main directory and run the following:
 
 ```
-mkdir -p build
-cd build
 @call "C:\Program Files (x86)\Intel\oneAPI\setvars.bat" intel64 --force
 
 # Option 1: Use FP32 (recommended for better performance in most cases)
-cmake -G "MinGW Makefiles" ..  -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icx  -DCMAKE_BUILD_TYPE=Release
+cmake -B build -G "MinGW Makefiles" -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icx  -DCMAKE_BUILD_TYPE=Release
 
 # Option 2: Or FP16
-cmake -G "MinGW Makefiles" ..  -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icx  -DCMAKE_BUILD_TYPE=Release -DLLAMA_SYCL_F16=ON
+cmake -B build -G "MinGW Makefiles" -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icx  -DCMAKE_BUILD_TYPE=Release -DLLAMA_SYCL_F16=ON
 
-make -j
+cmake --build build --config Release -j
 ```
 
 Otherwise, run the `win-build-sycl.bat` wrapper which encapsulates the former instructions:
diff --git a/README.md b/README.md
index a2aa9214f..514ef3af1 100644
--- a/README.md
+++ b/README.md
@@ -308,6 +308,8 @@ In order to build llama.cpp you have three different options.
       make
       ```
 
+      **Note**: for `Debug` builds, run `make LLAMA_DEBUG=1`
+
   - On Windows:
 
     1. Download the latest fortran version of [w64devkit](https://github.com/skeeto/w64devkit/releases).
@@ -322,12 +324,26 @@ In order to build llama.cpp you have three different options.
 - Using `CMake`:
 
     ```bash
-    mkdir build
-    cd build
-    cmake ..
-    cmake --build . --config Release
+    cmake -B build
+    cmake --build build --config Release
     ```
 
+    **Note**: for `Debug` builds, there are two cases:
+
+    - Single-config generators (e.g. default = `Unix Makefiles`; note that they just ignore the `--config` flag):
+
+      ```bash
+      cmake -B build -DCMAKE_BUILD_TYPE=Debug
+      cmake --build build
+      ```
+
+    - Multi-config generators (`-G` param set to Visual Studio, XCode...):
+
+      ```bash
+      cmake -B build -G "Xcode"
+      cmake --build build --config Debug
+      ```
+
 - Using `Zig` (version 0.11 or later):
 
     Building for optimization levels and CPU features can be accomplished using standard build arguments, for example AVX2, FMA, F16C,
@@ -439,10 +455,8 @@ Building the program with BLAS support may lead to some performance improvements
   - Using `CMake` on Linux:
 
       ```bash
-      mkdir build
-      cd build
-      cmake .. -DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS
-      cmake --build . --config Release
+      cmake -B build -DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS
+      cmake --build build --config Release
       ```
 
 - #### BLIS
@@ -462,11 +476,9 @@ Building the program with BLAS support may lead to some performance improvements
   - Using manual oneAPI installation:
     By default, `LLAMA_BLAS_VENDOR` is set to `Generic`, so if you already sourced intel environment script and assign `-DLLAMA_BLAS=ON` in cmake, the mkl version of Blas will automatically been selected. Otherwise please install oneAPI and follow the below steps:
       ```bash
-      mkdir build
-      cd build
       source /opt/intel/oneapi/setvars.sh # You can skip this step if  in oneapi-basekit docker image, only required for manual installation
-      cmake .. -DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=Intel10_64lp -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_NATIVE=ON
-      cmake --build . --config Release
+      cmake -B build -DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=Intel10_64lp -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_NATIVE=ON
+      cmake --build build --config Release
       ```
 
   - Using oneAPI docker image:
@@ -487,10 +499,8 @@ Building the program with BLAS support may lead to some performance improvements
   - Using `CMake`:
 
     ```bash
-    mkdir build
-    cd build
-    cmake .. -DLLAMA_CUDA=ON
-    cmake --build . --config Release
+    cmake -B build -DLLAMA_CUDA=ON
+    cmake --build build --config Release
     ```
 
   The environment variable [`CUDA_VISIBLE_DEVICES`](https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#env-vars) can be used to specify which GPU(s) will be used. The following compilation options are also available to tweak performance:
@@ -517,8 +527,8 @@ Building the program with BLAS support may lead to some performance improvements
   - Using `CMake` for Linux (assuming a gfx1030-compatible AMD GPU):
     ```bash
     CC=/opt/rocm/llvm/bin/clang CXX=/opt/rocm/llvm/bin/clang++ \
-        cmake -H. -Bbuild -DLLAMA_HIPBLAS=ON -DAMDGPU_TARGETS=gfx1030 -DCMAKE_BUILD_TYPE=Release \
-        && cmake --build build -- -j 16
+        cmake -B build -DLLAMA_HIPBLAS=ON -DAMDGPU_TARGETS=gfx1030 -DCMAKE_BUILD_TYPE=Release \
+        && cmake --build build --config Release -- -j 16
     ```
     On Linux it is also possible to use unified memory architecture (UMA) to share main memory between the CPU and integrated GPU by setting `-DLLAMA_HIP_UMA=ON"`.
     However, this hurts performance for non-integrated GPUs (but enables working with integrated GPUs).
@@ -564,15 +574,14 @@ Building the program with BLAS support may lead to some performance improvements
 
         ```sh
         git clone --recurse-submodules https://github.com/KhronosGroup/OpenCL-SDK.git
-        mkdir OpenCL-SDK/build
-        cd OpenCL-SDK/build
-        cmake .. -DBUILD_DOCS=OFF \
+        cd OpenCL-SDK
+        cmake -B build -DBUILD_DOCS=OFF \
           -DBUILD_EXAMPLES=OFF \
           -DBUILD_TESTING=OFF \
           -DOPENCL_SDK_BUILD_SAMPLES=OFF \
           -DOPENCL_SDK_TEST_SAMPLES=OFF
-        cmake --build . --config Release
-        cmake --install . --prefix /some/path
+        cmake --build build
+        cmake --install build --prefix /some/path
         ```
       </details>
 
@@ -594,23 +603,23 @@ Building the program with BLAS support may lead to some performance improvements
       ```cmd
       set OPENCL_SDK_ROOT="C:/OpenCL-SDK-v2023.04.17-Win-x64"
       git clone https://github.com/CNugteren/CLBlast.git
-      mkdir CLBlast\build
-      cd CLBlast\build
-      cmake .. -DBUILD_SHARED_LIBS=OFF -DOVERRIDE_MSVC_FLAGS_TO_MT=OFF -DTUNERS=OFF -DOPENCL_ROOT=%OPENCL_SDK_ROOT% -G "Visual Studio 17 2022" -A x64
-      cmake --build . --config Release
-      cmake --install . --prefix C:/CLBlast
+      cd CLBlast
+      cmake -B build -DBUILD_SHARED_LIBS=OFF -DOVERRIDE_MSVC_FLAGS_TO_MT=OFF -DTUNERS=OFF -DOPENCL_ROOT=%OPENCL_SDK_ROOT% -G "Visual Studio 17 2022" -A x64
+      cmake --build build --config Release
+      cmake --install build --prefix C:/CLBlast
       ```
 
+      (note: `--config Release` at build time is the default and only relevant for Visual Studio builds - or multi-config Ninja builds)
+
   - <details>
     <summary>Unix:</summary>
 
       ```sh
       git clone https://github.com/CNugteren/CLBlast.git
-      mkdir CLBlast/build
-      cd CLBlast/build
-      cmake .. -DBUILD_SHARED_LIBS=OFF -DTUNERS=OFF
-      cmake --build . --config Release
-      cmake --install . --prefix /some/path
+      cd CLBlast
+      cmake -B build -DBUILD_SHARED_LIBS=OFF -DTUNERS=OFF
+      cmake --build build --config Release
+      cmake --install build --prefix /some/path
       ```
 
       Where `/some/path` is where the built library will be installed (default is `/usr/local`).
@@ -624,21 +633,17 @@ Building the program with BLAS support may lead to some performance improvements
     ```
   - CMake (Unix):
     ```sh
-    mkdir build
-    cd build
-    cmake .. -DLLAMA_CLBLAST=ON -DCLBlast_DIR=/some/path
-    cmake --build . --config Release
+    cmake -B build -DLLAMA_CLBLAST=ON -DCLBlast_DIR=/some/path
+    cmake --build build --config Release
     ```
   - CMake (Windows):
     ```cmd
     set CL_BLAST_CMAKE_PKG="C:/CLBlast/lib/cmake/CLBlast"
     git clone https://github.com/ggerganov/llama.cpp
     cd llama.cpp
-    mkdir build
-    cd build
-    cmake .. -DBUILD_SHARED_LIBS=OFF -DLLAMA_CLBLAST=ON -DCMAKE_PREFIX_PATH=%CL_BLAST_CMAKE_PKG% -G "Visual Studio 17 2022" -A x64
-    cmake --build . --config Release
-    cmake --install . --prefix C:/LlamaCPP
+    cmake -B build -DBUILD_SHARED_LIBS=OFF -DLLAMA_CLBLAST=ON -DCMAKE_PREFIX_PATH=%CL_BLAST_CMAKE_PKG% -G "Visual Studio 17 2022" -A x64
+    cmake --build build --config Release
+    cmake --install build --prefix C:/LlamaCPP
     ```
 
   ##### Running Llama with CLBlast
@@ -694,10 +699,8 @@ Building the program with BLAS support may lead to some performance improvements
   Then, build llama.cpp using the cmake command below:
 
   ```bash
-  mkdir -p build
-  cd build
-  cmake .. -DLLAMA_VULKAN=1
-  cmake --build . --config Release
+  cmake -B build -DLLAMA_VULKAN=1
+  cmake --build build --config Release
   # Test the output binary (with "-ngl 33" to offload all layers to GPU)
   ./bin/main -m "PATH_TO_MODEL" -p "Hi you how are you" -n 50 -e -ngl 33 -t 4
 
diff --git a/examples/main-cmake-pkg/README.md b/examples/main-cmake-pkg/README.md
index f599fbaec..edf20d8db 100644
--- a/examples/main-cmake-pkg/README.md
+++ b/examples/main-cmake-pkg/README.md
@@ -17,11 +17,9 @@ In this case, CLBlast was already installed so the CMake package is referenced i
 ```cmd
 git clone https://github.com/ggerganov/llama.cpp
 cd llama.cpp
-mkdir build
-cd build
-cmake .. -DBUILD_SHARED_LIBS=OFF -DLLAMA_CLBLAST=ON -DCMAKE_PREFIX_PATH=C:/CLBlast/lib/cmake/CLBlast -G "Visual Studio 17 2022" -A x64
-cmake --build . --config Release
-cmake --install . --prefix C:/LlamaCPP
+cmake -B build -DBUILD_SHARED_LIBS=OFF -DLLAMA_CLBLAST=ON -DCMAKE_PREFIX_PATH=C:/CLBlast/lib/cmake/CLBlast -G "Visual Studio 17 2022" -A x64
+cmake --build build --config Release
+cmake --install build --prefix C:/LlamaCPP
 ```
 
 ### Build main-cmake-pkg
@@ -29,9 +27,7 @@ cmake --install . --prefix C:/LlamaCPP
 
 ```cmd
 cd ..\examples\main-cmake-pkg
-mkdir build
-cd build
-cmake .. -DBUILD_SHARED_LIBS=OFF -DCMAKE_PREFIX_PATH="C:/CLBlast/lib/cmake/CLBlast;C:/LlamaCPP/lib/cmake/Llama" -G "Visual Studio 17 2022" -A x64
-cmake --build . --config Release
-cmake --install . --prefix C:/MyLlamaApp
+cmake -B build -DBUILD_SHARED_LIBS=OFF -DCMAKE_PREFIX_PATH="C:/CLBlast/lib/cmake/CLBlast;C:/LlamaCPP/lib/cmake/Llama" -G "Visual Studio 17 2022" -A x64
+cmake --build build --config Release
+cmake --install build --prefix C:/MyLlamaApp
 ```
diff --git a/examples/server/README.md b/examples/server/README.md
index 918ac1295..b96a4444a 100644
--- a/examples/server/README.md
+++ b/examples/server/README.md
@@ -74,15 +74,18 @@ page cache before using this. See https://github.com/ggerganov/llama.cpp/issues/
 - Using `make`:
 
   ```bash
-  make
+  make server
   ```
 
 - Using `CMake`:
 
   ```bash
-  cmake --build . --config Release
+  cmake -B build
+  cmake --build build --config Release -t server
   ```
 
+  Binary is at `./build/bin/server`
+
 ## Build with SSL
 
 `server` can also be built with SSL support using OpenSSL 3
@@ -99,10 +102,8 @@ page cache before using this. See https://github.com/ggerganov/llama.cpp/issues/
 - Using `CMake`:
 
   ```bash
-  mkdir build
-  cd build
-  cmake .. -DLLAMA_SERVER_SSL=ON
-  make server
+  cmake -B build -DLLAMA_SERVER_SSL=ON
+  cmake --build build --config Release -t server
   ```
 
 ## Quick Start