Support unaligned M

.circleci/config.yml

@@ -7,9 +7,6 @@ parameters:
   nightly_build:
     type: boolean
     default: false
-  test_release:
-    type: boolean
-    default: false
 
 jobs:
   build_documentation:
@@ -81,24 +78,23 @@ jobs:
             export DEBIAN_FRONTEND=noninteractive
             export NEEDRESTART_MODE=a
             sudo apt-get update
+            sudo apt-get upgrade -y
+            pip install --upgrade cmake
             sudo apt-get install -y libblas-dev liblapack-dev liblapacke-dev
             sudo apt-get install openmpi-bin openmpi-common libopenmpi-dev
-            curl -LsSf https://astral.sh/uv/install.sh | sh
       - run:
           name: Install Python package
           command: |
-            uv venv
-            uv pip install cmake
-            uv pip install -e ".[dev]" -v
+            pip install -e ".[dev]"
       - run:
           name: Generate package stubs
           command: |
-            uv pip install typing_extensions
-            uv run --no-project setup.py generate_stubs
+            echo "stubs"
+            pip install typing_extensions
+            python setup.py generate_stubs
       - run:
           name: Run Python tests
           command: |
-            source .venv/bin/activate
             python -m unittest discover python/tests -v
             mpirun --bind-to none -host localhost:8 -np 8 python python/tests/mpi_test_distributed.py
             mlx.launch --verbose -n 8 python/tests/ring_test_distributed.py -v 2> >(tee -a stderr.log >&2)
@@ -106,7 +102,6 @@ jobs:
       - run:
           name: Build CPP only
           command: |
-            source .venv/bin/activate
             mkdir -p build && cd build
             cmake .. -DMLX_BUILD_METAL=OFF -DCMAKE_BUILD_TYPE=DEBUG
             make -j `nproc`
@@ -132,30 +127,33 @@ jobs:
       - run:
           name: Install dependencies
           command: |
-            HOMEBREW_NO_AUTO_UPDATE=1 HOMEBREW_NO_INSTALL_CLEANUP=1 \
-              brew install openmpi uv
+            brew install python@3.9
+            brew install openmpi
+            python3.9 -m venv env
+            source env/bin/activate
+            pip install --upgrade pip
+            pip install --upgrade cmake
+            pip install nanobind==2.4.0
+            pip install numpy
+            pip install torch
+            pip install tensorflow
+            pip install unittest-xml-reporting
       - run:
           name: Install Python package
           command: |
-            uv venv --python 3.9
-            uv pip install \
-              nanobind==2.4.0 \
-              cmake \
-              numpy \
-              torch \
-              tensorflow \
-              unittest-xml-reporting
+            source env/bin/activate
             DEBUG=1 CMAKE_ARGS="-DCMAKE_COMPILE_WARNING_AS_ERROR=ON" \
-              uv pip install -e . -v
+              pip install -e . -v
       - run:
           name: Generate package stubs
           command: |
-            uv pip install typing_extensions
-            uv run --no-project setup.py generate_stubs
+            source env/bin/activate
+            pip install typing_extensions
+            python setup.py generate_stubs
       - run:
           name: Run Python tests
           command: |
-            source .venv/bin/activate
+            source env/bin/activate
             LOW_MEMORY=1 DEVICE=cpu python -m xmlrunner discover -v python/tests -o test-results/cpu
             LOW_MEMORY=1 DEVICE=gpu METAL_DEVICE_WRAPPER_TYPE=1 METAL_DEBUG_ERROR_MODE=0 python -m xmlrunner discover -v python/tests -o test-results/gpu
             mpirun --bind-to none -host localhost:8 -np 8 -x DYLD_LIBRARY_PATH=/opt/homebrew/lib/ python python/tests/mpi_test_distributed.py
@@ -164,17 +162,16 @@ jobs:
       - run:
           name: Build example extension
           command: |
-            source .venv/bin/activate
+            source env/bin/activate
             cd examples/extensions
-            uv pip install -r requirements.txt
-            uv run --no-project setup.py build_ext --inplace
-            uv run --no-project python test.py
+            pip install -r requirements.txt
+            python setup.py build_ext -j8
       - store_test_results:
           path: test-results
       - run:
           name: Build CPP only
           command: |
-            source .venv/bin/activate
+            source env/bin/activate
             mkdir -p build && cd build && cmake .. && make -j `sysctl -n hw.ncpu`
       - run:
           name: Run CPP tests
@@ -183,7 +180,7 @@ jobs:
       - run:
           name: Build small binary
           command: |
-            source .venv/bin/activate
+            source env/bin/activate
             cd build/
             cmake .. -DCMAKE_BUILD_TYPE=MinSizeRel \
               -DBUILD_SHARED_LIBS=ON \
@@ -195,60 +192,34 @@ jobs:
       - run:
           name: Run Python tests with JIT
           command: |
+            source env/bin/activate
             CMAKE_ARGS="-DMLX_METAL_JIT=ON" \
-              uv pip install -e .
+              pip install -e . -v
             LOW_MEMORY=1 DEVICE=gpu METAL_DEVICE_WRAPPER_TYPE=1 \
               METAL_DEBUG_ERROR_MODE=0 \
-              uv run --no-project python -m xmlrunner discover \
-                -v python/tests \
-                -o test-results/gpu_jit
+              python -m xmlrunner discover -v python/tests -o test-results/gpu_jit
 
   cuda_build_and_test:
-    parameters:
-      image_date:
-        type: string
-        default: "2023.11.1"
     machine:
-      image: "linux-cuda-12:<< parameters.image_date >>"
+      image: linux-cuda-12:2023.11.1
       resource_class: gpu.nvidia.small.gen2
     steps:
       - checkout
-      - restore_cache:
-          keys:
-            - cuda-<< parameters.image_date >>-{{ arch }}-
-      - run:
-          name: Install dependencies
-          command: |
-            sudo apt-get update
-            sudo apt-get install libcudnn9-dev-cuda-12
-            sudo apt-get install libblas-dev liblapack-dev liblapacke-dev
-            curl -sL https://github.com/ccache/ccache/releases/download/v4.11.3/ccache-4.11.3-linux-x86_64.tar.xz | tar xJf -
-            sudo mv ccache-4.11.3-linux-x86_64/ccache /usr/bin/ccache
-            rm -rf ccache-4.11.3-linux-x86_64
-            curl -LsSf https://astral.sh/uv/install.sh | sh
       - run:
           name: Install Python package
           command: |
-            uv venv
+            sudo apt-get update
+            sudo apt-get install libblas-dev liblapack-dev liblapacke-dev
+            python3 -m venv env
+            source env/bin/activate
             CMAKE_ARGS="-DMLX_BUILD_CUDA=ON -DCMAKE_CUDA_COMPILER=`which nvcc`" \
-              uv pip install -e ".[dev]" -v
+              pip install -e ".[dev]"
       - run:
           name: Run Python tests
           command: |
-            source .venv/bin/activate
+            source env/bin/activate
             LOW_MEMORY=1 DEVICE=cpu python -m unittest discover python/tests -v
             LOW_MEMORY=1 DEVICE=gpu python -m tests discover python/tests -v
-      - run:
-          name: CCache report
-          command: |
-            ccache --show-stats
-            ccache --zero-stats
-            ccache --max-size 400MB
-            ccache --cleanup
-      - save_cache:
-          key: cuda-<< parameters.image_date >>-{{ arch }}-{{ epoch }}
-          paths:
-            - /home/circleci/.cache/ccache
 
   build_release:
     parameters:
@@ -344,10 +315,14 @@ jobs:
             export DEBIAN_FRONTEND=noninteractive
             export NEEDRESTART_MODE=a
             sudo apt-get update
+            sudo apt-get upgrade -y
             TZ=Etc/UTC sudo apt-get -y install tzdata
+            sudo apt-get install -y apt-utils
+            sudo apt-get install -y software-properties-common
             sudo add-apt-repository -y ppa:deadsnakes/ppa
             sudo apt-get install -y $PYTHON $PYTHON-dev $PYTHON-full
             sudo apt-get install -y libblas-dev liblapack-dev liblapacke-dev
+            sudo apt-get install -y build-essential git
             $PYTHON -m venv env
             source env/bin/activate
             pip install --upgrade pip
@@ -391,27 +366,22 @@ jobs:
         type: string
         default: ""
     machine:
-      image: ubuntu-2204:current
-      resource_class: large
+      image: linux-cuda-12:2024.11.1
+      resource_class: gpu.nvidia.small.gen2
     steps:
       - checkout
       - run:
           name: Build wheel
           command: |
-            export DEBIAN_FRONTEND=noninteractive
-            export NEEDRESTART_MODE=a
-            wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2404/x86_64/cuda-keyring_1.1-1_all.deb
-            sudo dpkg -i cuda-keyring_1.1-1_all.deb
             sudo apt-get update
-            sudo apt-get install cuda-toolkit-12-9 libcudnn9-dev-cuda-12
             sudo apt-get install libblas-dev liblapack-dev liblapacke-dev
             sudo apt-get install zip
+            python -m venv env
+            source env/bin/activate
             pip install auditwheel
             pip install patchelf
             pip install build
             pip install twine
-            export PATH=/usr/local/cuda/bin${PATH:+:${PATH}}
-            export LD_LIBRARY_PATH=/usr/local/cuda/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
             << parameters.build_env >> MLX_BUILD_STAGE=2 \
               CMAKE_ARGS="-DMLX_BUILD_CUDA=ON -DCMAKE_CUDA_COMPILER=`which nvcc`" \
               python -m build -w
@@ -422,6 +392,7 @@ jobs:
             - run:
                 name: Upload package
                 command: |
+                  source env/bin/activate
                   twine upload wheelhouse/*.whl
       - store_artifacts:
           path: wheelhouse/
@@ -434,24 +405,19 @@ workflows:
             pattern: "^(?!pull/)[-\\w]+$"
             value: << pipeline.git.branch >>
         - not: << pipeline.parameters.nightly_build >>
-        - not: << pipeline.parameters.test_release >>
     jobs:
       - mac_build_and_test:
           matrix:
             parameters:
               macosx_deployment_target: ["13.5", "14.0"]
       - linux_build_and_test
-      - cuda_build_and_test:
-          matrix:
-            parameters:
-              image_date: ["2023.11.1", "2025.05.1"]
+      - cuda_build_and_test 
       - build_documentation 
 
   build_pypi_release:
     when:
       and:
         - not: << pipeline.parameters.nightly_build >>
-        - not: << pipeline.parameters.test_release >>
     jobs:
       - build_release:
           filters:
@@ -572,9 +538,6 @@ workflows:
           requires: [ hold ]
       - cuda_build_and_test:
           requires: [ hold ]
-          matrix:
-            parameters:
-              image_date: ["2023.11.1", "2025.05.1"]
   nightly_build:
     when:
       and:
@@ -638,87 +601,3 @@ workflows:
             parameters:
               python_version: ["3.9", "3.10", "3.11", "3.12", "3.13"]
       - build_cuda_release
-
-  build_dev_release:
-    when:
-      and:
-        - equal: [ main, << pipeline.git.branch >> ]
-        - << pipeline.parameters.test_release >>
-    jobs:
-      - build_release:
-          matrix:
-            parameters:
-              python_version: ["3.9", "3.10", "3.11", "3.12", "3.13"]
-              macosx_deployment_target: ["13.5", "14.0", "15.0"]
-              build_env: ["DEV_RELEASE=1"]
-              xcode_version: ["16.2.0", "15.0.0"]
-            exclude:
-              - macosx_deployment_target: "13.5"
-                xcode_version: "16.2.0"
-                python_version: "3.9"
-                build_env: "DEV_RELEASE=1"
-              - macosx_deployment_target: "13.5"
-                xcode_version: "16.2.0"
-                python_version: "3.10"
-                build_env: "DEV_RELEASE=1"
-              - macosx_deployment_target: "13.5"
-                xcode_version: "16.2.0"
-                python_version: "3.11"
-                build_env: "DEV_RELEASE=1"
-              - macosx_deployment_target: "13.5"
-                xcode_version: "16.2.0"
-                python_version: "3.12"
-                build_env: "DEV_RELEASE=1"
-              - macosx_deployment_target: "13.5"
-                xcode_version: "16.2.0"
-                python_version: "3.13"
-                build_env: "DEV_RELEASE=1"
-              - macosx_deployment_target: "14.0"
-                xcode_version: "15.0.0"
-                python_version: "3.9"
-                build_env: "DEV_RELEASE=1"
-              - macosx_deployment_target: "14.0"
-                xcode_version: "15.0.0"
-                python_version: "3.10"
-                build_env: "DEV_RELEASE=1"
-              - macosx_deployment_target: "14.0"
-                xcode_version: "15.0.0"
-                python_version: "3.11"
-                build_env: "DEV_RELEASE=1"
-              - macosx_deployment_target: "14.0"
-                xcode_version: "15.0.0"
-                python_version: "3.12"
-                build_env: "DEV_RELEASE=1"
-              - macosx_deployment_target: "14.0"
-                xcode_version: "15.0.0"
-                python_version: "3.13"
-                build_env: "DEV_RELEASE=1"
-              - macosx_deployment_target: "15.0"
-                xcode_version: "15.0.0"
-                python_version: "3.9"
-                build_env: "DEV_RELEASE=1"
-              - macosx_deployment_target: "15.0"
-                xcode_version: "15.0.0"
-                python_version: "3.10"
-                build_env: "DEV_RELEASE=1"
-              - macosx_deployment_target: "15.0"
-                xcode_version: "15.0.0"
-                python_version: "3.11"
-                build_env: "DEV_RELEASE=1"
-              - macosx_deployment_target: "15.0"
-                xcode_version: "15.0.0"
-                python_version: "3.12"
-                build_env: "DEV_RELEASE=1"
-              - macosx_deployment_target: "15.0"
-                xcode_version: "15.0.0"
-                python_version: "3.13"
-                build_env: "DEV_RELEASE=1"
-      - build_linux_release:
-          matrix:
-            parameters:
-              python_version: ["3.9", "3.10", "3.11", "3.12", "3.13"]
-              build_env: ["DEV_RELEASE=1"]
-      - build_cuda_release:
-          matrix:
-            parameters:
-              build_env: ["DEV_RELEASE=1"]

CMakeLists.txt

@@ -22,7 +22,7 @@ project(
 
 # ----------------------------- Setup -----------------------------
 set(CMAKE_MODULE_PATH "${PROJECT_SOURCE_DIR}/cmake")
-set(CMAKE_CXX_STANDARD 17)
+set(CMAKE_CXX_STANDARD 20)
 set(CMAKE_CXX_STANDARD_REQUIRED ON)
 set(CMAKE_POSITION_INDEPENDENT_CODE ON)
 set(CMAKE_INSTALL_MESSAGE NEVER)
@@ -41,9 +41,7 @@ option(MLX_BUILD_GGUF "Include support for GGUF format" ON)
 option(MLX_BUILD_SAFETENSORS "Include support for safetensors format" ON)
 option(MLX_BUILD_BLAS_FROM_SOURCE "Build OpenBLAS from source code" OFF)
 option(MLX_METAL_JIT "Use JIT compilation for Metal kernels" OFF)
-option(MLX_USE_CCACHE "Use CCache for compilation cache when available" ON)
 option(BUILD_SHARED_LIBS "Build mlx as a shared library" OFF)
-option(USE_SYSTEM_FMT "Use system's provided fmt library" OFF)
 
 # --------------------- Processor tests -------------------------
 message(
@@ -70,15 +68,6 @@ else()
   set(MLX_BUILD_METAL OFF)
 endif()
 
-if(MLX_USE_CCACHE)
-  find_program(CCACHE_PROGRAM ccache)
-  if(CCACHE_PROGRAM)
-    set(CMAKE_C_COMPILER_LAUNCHER "${CCACHE_PROGRAM}")
-    set(CMAKE_CXX_COMPILER_LAUNCHER "${CCACHE_PROGRAM}")
-    set(CMAKE_CUDA_COMPILER_LAUNCHER "${CCACHE_PROGRAM}")
-  endif()
-endif()
-
 # ----------------------------- Lib -----------------------------
 
 include(FetchContent)
@@ -243,16 +232,12 @@ target_include_directories(
 # Do not add mlx_EXPORTS define for shared library.
 set_target_properties(mlx PROPERTIES DEFINE_SYMBOL "")
 
-if(USE_SYSTEM_FMT)
-  find_package(fmt REQUIRED)
-else()
-  FetchContent_Declare(
-    fmt
-    GIT_REPOSITORY https://github.com/fmtlib/fmt.git
-    GIT_TAG 10.2.1
-    EXCLUDE_FROM_ALL)
-  FetchContent_MakeAvailable(fmt)
-endif()
+FetchContent_Declare(
+  fmt
+  GIT_REPOSITORY https://github.com/fmtlib/fmt.git
+  GIT_TAG 10.2.1
+  EXCLUDE_FROM_ALL)
+FetchContent_MakeAvailable(fmt)
 target_link_libraries(mlx PRIVATE $<BUILD_INTERFACE:fmt::fmt-header-only>)
 
 if(MLX_BUILD_PYTHON_BINDINGS)

README.md

@@ -11,10 +11,10 @@ brought to you by Apple machine learning research.
 
 Some key features of MLX include:
 
- - **Familiar APIs**: MLX has a Python API that closely follows NumPy. MLX
+ - **Familiar APIs**: MLX has a Python API that closely follows NumPy.  MLX
    also has fully featured C++, [C](https://github.com/ml-explore/mlx-c), and
    [Swift](https://github.com/ml-explore/mlx-swift/) APIs, which closely mirror
-   the Python API. MLX has higher-level packages like `mlx.nn` and
+   the Python API.  MLX has higher-level packages like `mlx.nn` and
    `mlx.optimizers` with APIs that closely follow PyTorch to simplify building
    more complex models.
 
@@ -68,23 +68,18 @@ in the documentation.
 
 ## Installation
 
-MLX is available on [PyPI](https://pypi.org/project/mlx/). To install MLX on
-macOS, run:
+MLX is available on [PyPI](https://pypi.org/project/mlx/). To install the Python API, run:
 
-```bash
+**With `pip`**:
+
+```
 pip install mlx
 ```
 
-To install the CUDA backend on Linux, run:
+**With `conda`**:
 
-```bash
-pip install mlx[cuda]
 ```
-
-To install a CPU-only Linux package, run:
-
-```bash
-pip install mlx[cpu]
+conda install -c conda-forge mlx
 ```
 
 Checkout the

cmake/FindNCCL.cmake

@@ -1,54 +0,0 @@
-# FindNCCL.cmake This module finds the NVIDIA NCCL library and its include
-# directories.
-
-set(NCCL_ROOT_DIR
-    $ENV{NCCL_ROOT_DIR}
-    CACHE PATH "Folder contains NVIDIA NCCL")
-
-find_path(
-  NCCL_INCLUDE_DIRS
-  NAMES nccl.h
-  HINTS ${NCCL_INCLUDE_DIR} ${NCCL_ROOT_DIR} ${NCCL_ROOT_DIR}/include
-        ${CUDA_TOOLKIT_ROOT_DIR}/include)
-
-if($ENV{USE_STATIC_NCCL})
-  message(
-    STATUS "USE_STATIC_NCCL detected. Linking against static NCCL library")
-  set(NCCL_LIBNAME "libnccl_static.a")
-else()
-  set(NCCL_LIBNAME "nccl")
-endif()
-
-find_library(
-  NCCL_LIBRARIES
-  NAMES ${NCCL_LIBNAME}
-  HINTS ${NCCL_LIB_DIR}
-        ${NCCL_ROOT_DIR}
-        ${NCCL_ROOT_DIR}/lib
-        ${NCCL_ROOT_DIR}/lib/x86_64-linux-gnu
-        ${NCCL_ROOT_DIR}/lib64
-        ${CUDA_TOOLKIT_ROOT_DIR}/lib
-        ${CUDA_TOOLKIT_ROOT_DIR}/lib64)
-
-include(FindPackageHandleStandardArgs)
-find_package_handle_standard_args(NCCL DEFAULT_MSG NCCL_INCLUDE_DIRS
-                                  NCCL_LIBRARIES)
-
-if(NCCL_FOUND)
-  set(NCCL_HEADER_FILE "${NCCL_INCLUDE_DIRS}/nccl.h")
-  message(
-    STATUS "Determining NCCL version from the header file: ${NCCL_HEADER_FILE}")
-  file(
-    STRINGS ${NCCL_HEADER_FILE} NCCL_MAJOR_VERSION_DEFINED
-    REGEX "^[ \t]*#define[ \t]+NCCL_MAJOR[ \t]+[0-9]+.*$"
-    LIMIT_COUNT 1)
-  if(NCCL_MAJOR_VERSION_DEFINED)
-    string(REGEX REPLACE "^[ \t]*#define[ \t]+NCCL_MAJOR[ \t]+" ""
-                         NCCL_MAJOR_VERSION ${NCCL_MAJOR_VERSION_DEFINED})
-    message(STATUS "NCCL_MAJOR_VERSION: ${NCCL_MAJOR_VERSION}")
-  endif()
-  message(
-    STATUS
-      "Found NCCL (include: ${NCCL_INCLUDE_DIRS}, library: ${NCCL_LIBRARIES})")
-  mark_as_advanced(NCCL_ROOT_DIR NCCL_INCLUDE_DIRS NCCL_LIBRARIES)
-endif()

docs/src/dev/extensions.rst

@@ -394,14 +394,14 @@ below.
         out.set_data(allocator::malloc(out.nbytes()));
 
         // Resolve name of kernel
-        std::stream kname;
-        kname = "axpby_general_" + type_to_name(out);
+        std::ostringstream kname;
+        kname << "axpby_" << "general_" << type_to_name(out);
 
         // Load the metal library
-        auto lib = d.get_library("mlx_ext", current_binary_dir());
+        auto lib = d.get_library("mlx_ext");
 
         // Make a kernel from this metal library
-        auto kernel = d.get_kernel(kname, lib);
+        auto kernel = d.get_kernel(kname.str(), lib);
 
         // Prepare to encode kernel
         auto& compute_encoder = d.get_command_encoder(s.index);

docs/src/install.rst

@@ -13,7 +13,7 @@ silicon computer is
 
     pip install mlx
 
-To install from PyPI your system must meet the following requirements:
+To install from PyPI you must meet the following requirements:
 
 - Using an M series chip (Apple silicon)
 - Using a native Python >= 3.9
@@ -26,21 +26,12 @@ To install from PyPI your system must meet the following requirements:
 CUDA
 ^^^^
 
-MLX has a CUDA backend which you can install with:
+MLX has a CUDA backend which you can use on any Linux platform with CUDA 12
+and SM 7.0 (Volta) and up. To install MLX with CUDA support, run:
 
 .. code-block:: shell
 
-    pip install mlx[cuda]
-
-To install the CUDA package from PyPi your system must meet the following
-requirements:
-
-- Nvidia architecture >= SM 7.0 (Volta)
-- Nvidia driver >= 550.54.14
-- CUDA toolkit >= 12.0
-- Linux distribution with glibc >= 2.35
-- Python >= 3.9
-
+    pip install "mlx[cuda]"
 
 CPU-only (Linux)
 ^^^^^^^^^^^^^^^^
@@ -49,14 +40,7 @@ For a CPU-only version of MLX that runs on Linux use:
 
 .. code-block:: shell
 
-    pip install mlx[cpu]
-
-To install the CPU-only package from PyPi your system must meet the following
-requirements:
-
-- Linux distribution with glibc >= 2.35
-- Python >= 3.9
-
+    pip install "mlx[cpu]"
 
 Troubleshooting
 ^^^^^^^^^^^^^^^
@@ -271,7 +255,7 @@ and the CUDA toolkit. For example on Ubuntu, run the following:
    dpkg -i cuda-keyring_1.1-1_all.deb
    apt-get update -y
    apt-get -y install cuda-toolkit-12-9
-   apt-get install libblas-dev liblapack-dev liblapacke-dev libcudnn9-dev-cuda-12 -y
+   apt-get install libblas-dev liblapack-dev liblapacke-dev -y
 
 
 When building either the Python or C++ APIs make sure to pass the cmake flag

docs/src/python/optimizers.rst

@@ -51,14 +51,14 @@ the saved state. Here's a simple example:
    optimizer.update(model, grads)
 
    # Save the state
-   state = tree_flatten(optimizer.state, destination={})
-   mx.save_safetensors("optimizer.safetensors", state)
+   state = tree_flatten(optimizer.state)
+   mx.save_safetensors("optimizer.safetensors", dict(state))
 
    # Later on, for example when loading from a checkpoint,
    # recreate the optimizer and load the state
    optimizer = optim.Adam(learning_rate=1e-2)
 
-   state = tree_unflatten(mx.load("optimizer.safetensors"))
+   state = tree_unflatten(list(mx.load("optimizer.safetensors").items()))
    optimizer.state = state
 
 Note, not every optimizer configuation parameter is saved in the state. For

docs/src/usage/export.rst

@@ -7,17 +7,17 @@ Exporting Functions
 
 MLX has an API to export and import functions to and from a file. This lets you
 run computations written in one MLX front-end (e.g. Python) in another MLX
-front-end (e.g. C++).
+front-end (e.g. C++). 
 
 This guide walks through the basics of the MLX export API with some examples.
 To see the full list of functions check-out the :ref:`API documentation
 <export>`.
 
-Basics of Exporting
+Basics of Exporting 
 -------------------
 
 Let's start with a simple example:
-
+ 
 .. code-block:: python
 
   def fun(x, y):
@@ -67,7 +67,7 @@ specified as variable positional arguments or as a tuple of arrays:
 
   x = mx.array(1.0)
   y = mx.array(1.0)
-
+   
   # Both arguments to fun are positional
   mx.export_function("add.mlxfn", fun, x, y)
 
@@ -133,7 +133,7 @@ parameters are also saved to the ``model.mlxfn`` file.
    For enclosed arrays inside an exported function, be extra careful to ensure
    they are evaluated. The computation graph that gets exported will include
    the computation that produces enclosed inputs.
-
+  
    If the above example was missing ``mx.eval(model.parameters()``, the
    exported function would include the random initialization of the
    :obj:`mlx.nn.Module` parameters.
@@ -150,8 +150,8 @@ parameters, pass them as inputs to the ``call`` wrapper:
      # Set the model's parameters to the input parameters
      model.update(tree_unflatten(list(params.items())))
      return model(x)
-
-   params = tree_flatten(model.parameters(), destination={})
+ 
+   params = dict(tree_flatten(model.parameters()))
    mx.export_function("model.mlxfn", call, (mx.zeros(4),), params)
 
 
@@ -169,8 +169,8 @@ to export a function which can be used for inputs with variable shapes:
 
   # Ok
   out, = imported_abs(mx.array(-1.0))
-
-  # Also ok
+  
+  # Also ok 
   out, = imported_abs(mx.array([-1.0, -2.0]))
 
 With ``shapeless=False`` (which is the default), the second call to
@@ -197,7 +197,7 @@ a single file by creating an exporting context manager with :func:`exporter`:
   def fun(x, y=None):
       constant = mx.array(3.0)
       if y is not None:
-        x += y
+        x += y 
       return x + constant
 
   with mx.exporter("fun.mlxfn", fun) as exporter:
@@ -215,7 +215,7 @@ a single file by creating an exporting context manager with :func:`exporter`:
   print(out)
 
 In the above example the function constant data, (i.e. ``constant``), is only
-saved once.
+saved once. 
 
 Transformations with Imported Functions
 ---------------------------------------
@@ -238,7 +238,7 @@ on imported functions just like regular Python functions:
   # Prints: array(1, dtype=float32)
   print(dfdx(x))
 
-  # Compile the imported function
+  # Compile the imported function 
   mx.compile(imported_fun)
   # Prints: array(0, dtype=float32)
   print(compiled_fun(x)[0])
@@ -275,7 +275,7 @@ Import and run the function in C++ with only a few lines of code:
   // Prints: array(2, dtype=float32)
   std::cout << outputs[0] << std::endl;
 
-Imported functions can be transformed in C++ just like in Python. Use
+Imported functions can be transformed in C++ just like in Python. Use 
 ``std::vector<mx::array>`` for positional arguments and ``std::map<std::string,
 mx::array>`` for keyword arguments when calling imported functions in C++.
 

examples/extensions/axpby/axpby.cpp

@@ -1,6 +1,5 @@
 // Copyright © 2023-2025 Apple Inc.
 
-#include <dlfcn.h>
 #include <iostream>
 #include <sstream>
 
@@ -17,19 +16,6 @@
 
 namespace my_ext {
 
-// A helper function to find the location of the current binary on disk.
-// The Metal library ("mlx_ext.mtllib"), should be in the same directory.
-std::string current_binary_dir() {
-  static std::string binary_dir = []() {
-    Dl_info info;
-    if (!dladdr(reinterpret_cast<void*>(&current_binary_dir), &info)) {
-      throw std::runtime_error("Unable to get current binary dir.");
-    }
-    return std::filesystem::path(info.dli_fname).parent_path().string();
-  }();
-  return binary_dir;
-}
-
 ///////////////////////////////////////////////////////////////////////////////
 // Operation Implementation
 ///////////////////////////////////////////////////////////////////////////////
@@ -181,15 +167,16 @@ void Axpby::eval_gpu(
   }
 
   // Resolve name of kernel (corresponds to axpby.metal)
-  std::string kname = "axpby_";
-  kname += (contiguous_kernel ? "contiguous_" : "general_");
-  kname += type_to_name(out);
+  std::ostringstream kname;
+  kname << "axpby_";
+  kname << (contiguous_kernel ? "contiguous_" : "general_");
+  kname << type_to_name(out);
 
   // Load the metal library
-  auto lib = d.get_library("mlx_ext", current_binary_dir());
+  auto lib = d.get_library("mlx_ext");
 
   // Make a kernel from this metal library
-  auto kernel = d.get_kernel(kname, lib);
+  auto kernel = d.get_kernel(kname.str(), lib);
 
   // Prepare to encode kernel
   auto& compute_encoder = d.get_command_encoder(s.index);

examples/extensions/requirements.txt

@@ -1,4 +1,4 @@
 setuptools>=42
 cmake>=3.25
 mlx>=0.21.0
-nanobind==2.4.0
+nanobind==2.2.0

examples/extensions/test.py

@@ -3,10 +3,8 @@ from mlx_sample_extensions import axpby
 
 a = mx.ones((3, 4))
 b = mx.ones((3, 4))
-c_cpu = axpby(a, b, 4.0, 2.0, stream=mx.cpu)
-c_gpu = axpby(a, b, 4.0, 2.0, stream=mx.gpu)
+c = axpby(a, b, 4.0, 2.0, stream=mx.cpu)
 
-print(f"c shape: {c_cpu.shape}")
-print(f"c dtype: {c_cpu.dtype}")
-print(f"c_cpu correct: {mx.all(c_cpu == 6.0).item()}")
-print(f"c_gpu correct: {mx.all(c_gpu == 6.0).item()}")
+print(f"c shape: {c.shape}")
+print(f"c dtype: {c.dtype}")
+print(f"c correct: {mx.all(c == 6.0).item()}")

mlx/array.h

@@ -10,7 +10,6 @@
 #include "mlx/allocator.h"
 #include "mlx/dtype.h"
 #include "mlx/event.h"
-#include "mlx/small_vector.h"
 
 namespace mlx::core {
 
@@ -19,8 +18,8 @@ class Primitive;
 
 using Deleter = std::function<void(allocator::Buffer)>;
 using ShapeElem = int32_t;
-using Shape = SmallVector<ShapeElem>;
-using Strides = SmallVector<int64_t>;
+using Shape = std::vector<ShapeElem>;
+using Strides = std::vector<int64_t>;
 
 class array {
   /* An array is really a node in a graph. It contains a shared ArrayDesc

mlx/backend/common/utils.cpp

@@ -228,31 +228,4 @@ std::pair<Dims, Dims> get_grid_and_block_common(int dim0, int dim1, int dim2) {
       std::make_tuple(gx, gy, gz), std::make_tuple(bx, by, bz));
 }
 
-array swapaxes_in_eval(const array& x, int axis1, int axis2) {
-  int ndim = x.ndim();
-  if (axis1 < 0) {
-    axis1 += ndim;
-  }
-  if (axis2 < 0) {
-    axis2 += ndim;
-  }
-
-  auto shape = x.shape();
-  std::swap(shape[axis1], shape[axis2]);
-  auto strides = x.strides();
-  std::swap(strides[axis1], strides[axis2]);
-
-  auto [data_size, row_contiguous, col_contiguous] =
-      check_contiguity(shape, strides);
-  bool contiguous = data_size == x.data_size();
-
-  array out(std::move(shape), x.dtype(), nullptr, {});
-  out.copy_shared_buffer(
-      x,
-      std::move(strides),
-      {contiguous, row_contiguous, col_contiguous},
-      x.data_size());
-  return out;
-}
-
 } // namespace mlx::core

mlx/backend/common/utils.h

@@ -196,11 +196,8 @@ void shared_buffer_reshape(
     const Strides& out_strides,
     array& out);
 
-// Like the swapaxes op but safe to call in eval_gpu.
-array swapaxes_in_eval(const array& x, int axis1, int axis2);
-
 template <typename T>
-inline SmallVector<T> remove_index(SmallVector<T> vec, size_t index) {
+inline std::vector<T> remove_index(std::vector<T> vec, size_t index) {
   vec.erase(std::next(vec.begin(), index));
   return vec;
 }

mlx/backend/cpu/compiled.cpp

@@ -288,14 +288,6 @@ void Compiled::eval_cpu(
   auto [contiguous, shape, strides] =
       compiled_collapse_contiguous_dims(inputs, outputs[0], is_constant_);
 
-  // Force allocating shape/strides on heap so we can take their data() first
-  // and then std::move them.
-  // TODO: Refactor code to avoid heap allocation.
-  shape.grow();
-  for (auto& s : strides) {
-    s.grow();
-  }
-
   // Collect function input arguments.
   std::vector<void*> args;
   int strides_index = 1;

mlx/backend/cpu/copy.cpp

@@ -377,10 +377,4 @@ void copy_cpu_inplace(
       });
 }
 
-array contiguous_copy_cpu(const array& arr, Stream stream) {
-  array arr_copy(arr.shape(), arr.dtype(), nullptr, {});
-  copy_cpu(arr, arr_copy, CopyType::General, stream);
-  return arr_copy;
-}
-
 } // namespace mlx::core

mlx/backend/cpu/copy.h

@@ -30,7 +30,4 @@ void copy_cpu_inplace(
     const std::optional<array>& dynamic_i_offset = std::nullopt,
     const std::optional<array>& dynamic_o_offset = std::nullopt);
 
-// Return a contiguous array with same shape that copies the data of |arr|.
-array contiguous_copy_cpu(const array& arr, Stream stream);
-
 } // namespace mlx::core

mlx/backend/cpu/distributed.cpp

@@ -13,7 +13,9 @@ std::pair<array, bool> ensure_row_contiguous(const array& arr, Stream stream) {
   if (arr.flags().row_contiguous) {
     return {arr, false};
   } else {
-    return {contiguous_copy_cpu(arr, stream), true};
+    array arr_copy(arr.shape(), arr.dtype(), nullptr, {});
+    copy_cpu(arr, arr_copy, CopyType::General, stream);
+    return {arr_copy, true};
   }
 };
 
@@ -32,7 +34,8 @@ void AllReduce::eval_cpu(
       }
       return in;
     } else {
-      array arr_copy = contiguous_copy_cpu(in, s);
+      array arr_copy(in.shape(), in.dtype(), nullptr, {});
+      copy_cpu(in, arr_copy, CopyType::General, s);
       out.copy_shared_buffer(arr_copy);
       return arr_copy;
     }

mlx/backend/cpu/jit_compiler.cpp

@@ -2,7 +2,6 @@
 
 #include "mlx/backend/cpu/jit_compiler.h"
 
-#include <algorithm>
 #include <sstream>
 #include <vector>
 

mlx/backend/cpu/logsumexp.cpp

@@ -87,7 +87,8 @@ void LogSumExp::eval_cpu(const std::vector<array>& inputs, array& out) {
     if (x.flags().contiguous && x.strides()[x.ndim() - 1] == 1) {
       return x;
     } else {
-      array x_copy = contiguous_copy_cpu(x, s);
+      auto x_copy = array(x.shape(), x.dtype(), nullptr, {});
+      copy_cpu(x, x_copy, CopyType::General, s);
       encoder.add_temporary(x_copy);
       return x_copy;
     }

mlx/backend/cpu/masked_mm.cpp

@@ -136,8 +136,9 @@ void BlockMaskedMM::eval_cpu(const std::vector<array>& inputs, array& out) {
           }
           return std::make_tuple(true, sty, arr, false);
         } else {
+          array arr_copy(arr.shape(), arr.dtype(), nullptr, {});
+          copy_cpu(arr, arr_copy, CopyType::General, s);
           int64_t stx = arr.shape(-1);
-          array arr_copy = contiguous_copy_cpu(arr, s);
           return std::make_tuple(false, stx, arr_copy, true);
         }
       };

mlx/backend/cpu/quantized.cpp

@@ -712,7 +712,9 @@ void fast::AffineQuantize::eval_cpu(
     if (arr.flags().row_contiguous) {
       return std::make_pair(arr, false);
     } else {
-      return std::make_pair(contiguous_copy_cpu(arr, s), true);
+      array arr_copy(arr.shape(), arr.dtype(), nullptr, {});
+      copy_cpu(arr, arr_copy, CopyType::General, s);
+      return std::make_pair(arr_copy, true);
     }
   };
 

mlx/backend/cpu/scan.cpp

@@ -250,8 +250,10 @@ void Scan::eval_cpu(const std::vector<array>& inputs, array& out) {
   // Ensure contiguity
   auto in = inputs[0];
   if (!in.flags().row_contiguous) {
-    in = contiguous_copy_cpu(in, stream());
-    encoder.add_temporary(in);
+    array arr_copy(in.shape(), in.dtype(), nullptr, {});
+    copy_cpu(in, arr_copy, CopyType::General, stream());
+    in = arr_copy;
+    encoder.add_temporary(arr_copy);
   }
   out.set_data(allocator::malloc(out.nbytes()));
 

mlx/backend/cpu/softmax.cpp

@@ -131,7 +131,8 @@ void Softmax::eval_cpu(const std::vector<array>& inputs, array& out) {
       }
       return x;
     } else {
-      array x_copy = contiguous_copy_cpu(x, s);
+      array x_copy(x.shape(), x.dtype(), nullptr, {});
+      copy_cpu(x, x_copy, CopyType::General, s);
       out.copy_shared_buffer(x_copy);
       return x_copy;
     }

mlx/backend/cpu/sort.cpp

@@ -8,7 +8,7 @@
 #include "mlx/backend/common/utils.h"
 #include "mlx/backend/cpu/copy.h"
 #include "mlx/backend/cpu/encoder.h"
-#include "mlx/dtype_utils.h"
+
 #include "mlx/primitives.h"
 
 namespace mlx::core {
@@ -333,24 +333,47 @@ void Sort::eval_cpu(const std::vector<array>& inputs, array& out) {
   assert(inputs.size() == 1);
   auto& in = inputs[0];
 
-  int axis = axis_;
-  if (axis < 0) {
-    axis += in.ndim();
-  }
-
   // Copy input to output
-  CopyType ctype = (in.flags().contiguous && in.strides()[axis] != 0)
+  CopyType ctype = (in.flags().contiguous && in.strides()[axis_] != 0)
       ? CopyType::Vector
       : CopyType::General;
   copy_cpu(in, out, ctype, stream());
 
   auto& encoder = cpu::get_command_encoder(stream());
   encoder.set_output_array(out);
-  encoder.dispatch([out = array::unsafe_weak_copy(out), axis]() mutable {
-    dispatch_all_types(out.dtype(), [&](auto type_tag) {
-      sort<MLX_GET_TYPE(type_tag)>(out, axis);
-    });
-  });
+  encoder.dispatch(
+      [out = array::unsafe_weak_copy(out), axis_ = axis_]() mutable {
+        switch (out.dtype()) {
+          case bool_:
+            return sort<bool>(out, axis_);
+          case uint8:
+            return sort<uint8_t>(out, axis_);
+          case uint16:
+            return sort<uint16_t>(out, axis_);
+          case uint32:
+            return sort<uint32_t>(out, axis_);
+          case uint64:
+            return sort<uint64_t>(out, axis_);
+          case int8:
+            return sort<int8_t>(out, axis_);
+          case int16:
+            return sort<int16_t>(out, axis_);
+          case int32:
+            return sort<int32_t>(out, axis_);
+          case int64:
+            return sort<int64_t>(out, axis_);
+          case float32:
+            return sort<float>(out, axis_);
+          case float64:
+            return sort<double>(out, axis_);
+          case float16:
+            return sort<float16_t>(out, axis_);
+          case bfloat16:
+            return sort<bfloat16_t>(out, axis_);
+          case complex64:
+            return sort<complex64_t>(out, axis_);
+        }
+      });
 }
 
 void ArgPartition::eval_cpu(const std::vector<array>& inputs, array& out) {

mlx/backend/cuda/CMakeLists.txt

@@ -6,7 +6,6 @@
 target_sources(
   mlx
   PRIVATE ${CMAKE_CURRENT_SOURCE_DIR}/allocator.cpp
-          ${CMAKE_CURRENT_SOURCE_DIR}/arange.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/arg_reduce.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/binary.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/binary_two.cu
@@ -16,22 +15,18 @@ target_sources(
           ${CMAKE_CURRENT_SOURCE_DIR}/copy/copy_general.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/copy/copy_general_dynamic.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/copy/copy_general_input.cu
-          ${CMAKE_CURRENT_SOURCE_DIR}/conv.cpp
           ${CMAKE_CURRENT_SOURCE_DIR}/cuda.cpp
           ${CMAKE_CURRENT_SOURCE_DIR}/device.cpp
-          ${CMAKE_CURRENT_SOURCE_DIR}/distributed.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/eval.cpp
           ${CMAKE_CURRENT_SOURCE_DIR}/event.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/fence.cpp
-          ${CMAKE_CURRENT_SOURCE_DIR}/gemms/gemv.cu
-          ${CMAKE_CURRENT_SOURCE_DIR}/gemms/cublas_gemm.cpp
           ${CMAKE_CURRENT_SOURCE_DIR}/jit_module.cpp
           ${CMAKE_CURRENT_SOURCE_DIR}/indexing.cpp
           ${CMAKE_CURRENT_SOURCE_DIR}/kernel_utils.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/matmul.cpp
           ${CMAKE_CURRENT_SOURCE_DIR}/layer_norm.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/logsumexp.cu
-          ${CMAKE_CURRENT_SOURCE_DIR}/primitives.cpp
+          ${CMAKE_CURRENT_SOURCE_DIR}/primitives.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/random.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/reduce.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/reduce/all_reduce.cu
@@ -40,7 +35,6 @@ target_sources(
           ${CMAKE_CURRENT_SOURCE_DIR}/reduce/row_reduce.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/rms_norm.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/rope.cu
-          ${CMAKE_CURRENT_SOURCE_DIR}/scaled_dot_product_attention.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/scan.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/slicing.cpp
           ${CMAKE_CURRENT_SOURCE_DIR}/softmax.cu
@@ -49,17 +43,10 @@ target_sources(
           ${CMAKE_CURRENT_SOURCE_DIR}/unary.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/utils.cpp
           ${CMAKE_CURRENT_SOURCE_DIR}/quantized/affine_quantize.cu
-          ${CMAKE_CURRENT_SOURCE_DIR}/quantized/quantized.cpp
+          ${CMAKE_CURRENT_SOURCE_DIR}/quantized/qmm.cu
+          ${CMAKE_CURRENT_SOURCE_DIR}/quantized/quantized.cu
           ${CMAKE_CURRENT_SOURCE_DIR}/worker.cpp)
 
-if(CMAKE_CUDA_COMPILER_VERSION VERSION_GREATER_EQUAL 12.9.0)
-  target_sources(
-    mlx PRIVATE ${CMAKE_CURRENT_SOURCE_DIR}/gemms/cublas_batched_gemm_12_9.cu)
-else()
-  target_sources(
-    mlx PRIVATE ${CMAKE_CURRENT_SOURCE_DIR}/gemms/cublas_batched_gemm_12_0.cpp)
-endif()
-
 target_compile_definitions(mlx PRIVATE MLX_USE_CUDA)
 
 # Embed kernel sources in binary for JIT compilation.
@@ -102,18 +89,11 @@ endif()
 target_compile_options(
   mlx PRIVATE "$<$<COMPILE_LANGUAGE:CUDA>:--Wno-deprecated-gpu-targets>")
 
-# Use stronger binaries compression. This feature was introduced in CUDA 12.8
-# and requires drivers released after CUDA 12.4.
-if(CMAKE_CUDA_COMPILER_VERSION VERSION_GREATER_EQUAL 12.8.0)
-  target_compile_options(
-    mlx PRIVATE "$<$<COMPILE_LANGUAGE:CUDA>:--compress-mode=size>")
-endif()
-
-# Compute capability >= 7.0 is required for synchronization between CPU/GPU with
-# managed memory.
-if(NOT DEFINED MLX_CUDA_ARCHITECTURES)
-  set(MLX_CUDA_ARCHITECTURES "native")
-endif()
+# Compute capability 7 is required for synchronization between CPU/GPU with
+# managed memory. TODO: Add more architectures for potential performance gain.
+set(MLX_CUDA_ARCHITECTURES
+    "80"
+    CACHE STRING "CUDA architectures")
 message(STATUS "CUDA architectures: ${MLX_CUDA_ARCHITECTURES}")
 set_target_properties(mlx PROPERTIES CUDA_ARCHITECTURES
                                      "${MLX_CUDA_ARCHITECTURES}")
@@ -145,23 +125,6 @@ target_link_libraries(mlx PRIVATE CUDA::cublasLt)
 # Use NVRTC and driver APIs.
 target_link_libraries(mlx PRIVATE CUDA::nvrtc CUDA::cuda_driver)
 
-# Use the frontend APIs of cuDNN.
-FetchContent_Declare(
-  cudnn
-  GIT_REPOSITORY https://github.com/NVIDIA/cudnn-frontend.git
-  GIT_TAG v1.12.1
-  GIT_SHALLOW TRUE
-  EXCLUDE_FROM_ALL)
-set(CUDNN_FRONTEND_SKIP_JSON_LIB ON)
-set(CUDNN_FRONTEND_BUILD_SAMPLES OFF)
-set(CUDNN_FRONTEND_BUILD_TESTS OFF)
-set(CUDNN_FRONTEND_BUILD_PYTHON_BINDINGS OFF)
-FetchContent_MakeAvailable(cudnn)
-target_link_libraries(mlx PRIVATE cudnn_frontend)
-# Link with the actual cuDNN libraries.
-include(${cudnn_frontend_SOURCE_DIR}/cmake/cuDNN.cmake)
-target_link_libraries(mlx PRIVATE CUDNN::cudnn_all)
-
 # Suppress nvcc warnings on MLX headers.
 target_compile_options(mlx PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:-Xcudafe
                                    --diag_suppress=997>)
@@ -169,3 +132,12 @@ target_compile_options(mlx PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:-Xcudafe
 # Install CCCL headers for JIT.
 install(DIRECTORY ${cccl_SOURCE_DIR}/include/cuda
         DESTINATION ${CMAKE_INSTALL_INCLUDEDIR}/cccl)
+
+# Make Thunderkittens available
+FetchContent_Declare(
+  kittens
+  GIT_REPOSITORY https://github.com/HazyResearch/ThunderKittens.git
+  GIT_TAG aaab847f430ed313ed466e64b25b9177babd1db8
+  GIT_SHALLOW TRUE)
+FetchContent_MakeAvailable(kittens)
+target_include_directories(mlx BEFORE PRIVATE "${kittens_SOURCE_DIR}/include")

mlx/backend/cuda/allocator.cpp

@@ -2,6 +2,7 @@
 
 #include "mlx/backend/cuda/allocator.h"
 #include "mlx/backend/cuda/utils.h"
+#include "mlx/backend/cuda/worker.h"
 #include "mlx/utils.h"
 
 #include <cuda_runtime.h>
@@ -24,58 +25,52 @@ constexpr int small_block_size = 8;
 constexpr int small_pool_size = 4 * page_size;
 
 SmallSizePool::SmallSizePool() {
+  CHECK_CUDA_ERROR(cudaMallocManaged(&buffer_, small_pool_size));
+  end_ = reinterpret_cast<void*>(
+      reinterpret_cast<char*>(buffer_) + small_pool_size);
+  next_free_ = reinterpret_cast<Block*>(buffer_);
+
   auto num_blocks = small_pool_size / small_block_size;
-  buffer_ = new Block[num_blocks];
-
-  next_free_ = buffer_;
-
-  CHECK_CUDA_ERROR(cudaMallocManaged(&data_, small_pool_size));
-  CHECK_CUDA_ERROR(
-      cudaMemAdvise(data_, small_pool_size, cudaMemAdviseSetReadMostly, 0));
-
   auto curr = next_free_;
-  for (size_t i = 1; i < num_blocks; ++i) {
-    curr->next = buffer_ + i;
+  for (size_t i = 0; i < num_blocks - 1; ++i) {
+    curr->next = reinterpret_cast<Block*>(
+        reinterpret_cast<char*>(buffer_) + (i + 1) * small_block_size);
     curr = curr->next;
   }
   curr->next = nullptr;
 }
 
 SmallSizePool::~SmallSizePool() {
-  CHECK_CUDA_ERROR(cudaFree(data_));
-  delete[] buffer_;
+  CHECK_CUDA_ERROR(cudaFree(buffer_));
 }
 
-CudaBuffer* SmallSizePool::malloc() {
+void* SmallSizePool::malloc() {
   if (next_free_ == nullptr) {
     return nullptr;
   }
   Block* b = next_free_;
-  uint64_t i = next_free_ - buffer_;
   next_free_ = next_free_->next;
-  b->buf.data = static_cast<char*>(data_) + i * small_block_size;
-  b->buf.size = small_block_size;
-  return &b->buf;
+  return static_cast<void*>(b);
 }
 
-void SmallSizePool::free(CudaBuffer* buf) {
-  auto b = reinterpret_cast<Block*>(buf);
+void SmallSizePool::free(void* p) {
+  auto b = static_cast<Block*>(p);
   b->next = next_free_;
   next_free_ = b;
 }
 
-bool SmallSizePool::in_pool(CudaBuffer* buf) {
-  constexpr int num_blocks = (small_pool_size / small_block_size);
-  auto b = reinterpret_cast<Block*>(buf);
-  int64_t block_num = b - buffer_;
-  return block_num >= 0 && block_num < num_blocks;
+bool SmallSizePool::in_pool(void* p) {
+  return (p >= buffer_) && (p < end_);
 }
 
 CudaAllocator::CudaAllocator()
     : buffer_cache_(
           page_size,
           [](CudaBuffer* buf) { return buf->size; },
-          [this](CudaBuffer* buf) { cuda_free(buf); }) {
+          [this](CudaBuffer* buf) {
+            cuda_free(buf->data);
+            delete buf;
+          }) {
   // TODO: Set memory limit for multi-device.
   size_t free, total;
   CHECK_CUDA_ERROR(cudaMemGetInfo(&free, &total));
@@ -97,26 +92,28 @@ Buffer CudaAllocator::malloc(size_t size) {
 
   CudaBuffer* buf = buffer_cache_.reuse_from_cache(size);
   if (!buf) {
-    // If we have a lot of memory pressure try to reclaim memory from the cache.
-    int64_t mem_to_free =
-        get_active_memory() + get_cache_memory() + size - memory_limit_;
-    if (mem_to_free > 0) {
-      buffer_cache_.release_cached_buffers(mem_to_free);
+    // If we have a lot of memory pressure or are over the maximum cache size,
+    // try to reclaim memory from the cache.
+    size_t mem_required = get_active_memory() + get_cache_memory() + size;
+    if (mem_required >= memory_limit_) {
+      buffer_cache_.release_cached_buffers(mem_required - memory_limit_);
     }
 
+    lock.unlock();
+    buf = new CudaBuffer{nullptr, size};
+
     // Try the scalar pool first
     if (size <= small_block_size) {
-      buf = scalar_pool_.malloc();
+      buf->data = scalar_pool_.malloc();
     }
-    lock.unlock();
-    if (!buf) {
-      buf = new CudaBuffer{nullptr, size};
+    if (!buf->data) {
       cudaError_t err = cudaMallocManaged(&buf->data, size);
       if (err != cudaSuccess && err != cudaErrorMemoryAllocation) {
         throw std::runtime_error(fmt::format(
             "cudaMallocManaged failed: {}.", cudaGetErrorString(err)));
       }
     }
+
     lock.lock();
   }
   active_memory_ += size;
@@ -126,6 +123,7 @@ Buffer CudaAllocator::malloc(size_t size) {
   if (get_cache_memory() > max_pool_size_) {
     buffer_cache_.release_cached_buffers(get_cache_memory() - max_pool_size_);
   }
+
   return Buffer{buf};
 }
 
@@ -140,7 +138,9 @@ void CudaAllocator::free(Buffer buffer) {
   if (get_cache_memory() < max_pool_size_) {
     buffer_cache_.recycle_to_cache(buf);
   } else {
-    cuda_free(buf);
+    lock.unlock();
+    cuda_free(buf->data);
+    delete buf;
   }
 }
 
@@ -152,13 +152,30 @@ size_t CudaAllocator::size(Buffer buffer) const {
   return buf->size;
 }
 
-// This must be called with mutex_ aquired
-void CudaAllocator::cuda_free(CudaBuffer* buf) {
+void CudaAllocator::register_this_thread() {
+  std::lock_guard lock(worker_mutex_);
+  allowed_threads_.insert(std::this_thread::get_id());
+}
+
+void CudaAllocator::cuda_free(void* buf) {
+  // If cuda_free() is called from a unregistered thread, reschedule the call to
+  // worker.
+  {
+    std::lock_guard lock(worker_mutex_);
+    if (allowed_threads_.count(std::this_thread::get_id()) == 0) {
+      if (!worker_) {
+        worker_.reset(new Worker);
+      }
+      worker_->add_task([this, buf]() { this->cuda_free(buf); });
+      worker_->end_batch();
+      worker_->commit();
+      return;
+    }
+  }
   if (scalar_pool_.in_pool(buf)) {
     scalar_pool_.free(buf);
   } else {
-    cudaFree(buf->data);
-    delete buf;
+    cudaFree(buf);
   }
 }
 

mlx/backend/cuda/allocator.h

@@ -7,10 +7,13 @@
 
 #include <mutex>
 #include <set>
+#include <thread>
 #include <utility>
 
 namespace mlx::core::cu {
 
+class Worker;
+
 using allocator::Buffer;
 
 // Stores cuda-managed unified memory.
@@ -21,14 +24,13 @@ struct CudaBuffer {
 
 class SmallSizePool {
  private:
-  union Block {
+  struct Block {
     Block* next;
-    CudaBuffer buf;
   };
 
-  Block* buffer_{nullptr};
-  void* data_{nullptr};
+  void* buffer_{nullptr};
   Block* next_free_{nullptr};
+  void* end_{nullptr};
 
  public:
   SmallSizePool();
@@ -37,9 +39,9 @@ class SmallSizePool {
   SmallSizePool(const SmallSizePool&) = delete;
   SmallSizePool& operator=(const SmallSizePool&) = delete;
 
-  CudaBuffer* malloc();
-  void free(CudaBuffer* buf);
-  bool in_pool(CudaBuffer* buf);
+  void* malloc();
+  void free(void* p);
+  bool in_pool(void* p);
 };
 
 class CudaAllocator : public allocator::Allocator {
@@ -48,6 +50,15 @@ class CudaAllocator : public allocator::Allocator {
   void free(Buffer buffer) override;
   size_t size(Buffer buffer) const override;
 
+  // Register current thread as safe to free buffers.
+  // In cuda freeing a buffer implicitly synchronizes stream, and for threads
+  // that may be waited by gpu stream (for example cpu stream threads), freeing
+  // buffers there would result in dead lock.
+  void register_this_thread();
+
+  // Call cudaFree in the safe thread.
+  void cuda_free(void* buf);
+
   size_t get_active_memory() const;
   size_t get_peak_memory() const;
   void reset_peak_memory();
@@ -58,11 +69,13 @@ class CudaAllocator : public allocator::Allocator {
   void clear_cache();
 
  private:
-  void cuda_free(CudaBuffer* buf);
-
   CudaAllocator();
   friend CudaAllocator& allocator();
 
+  std::mutex worker_mutex_;
+  std::unique_ptr<Worker> worker_;
+  std::set<std::thread::id> allowed_threads_;
+
   std::mutex mutex_;
   size_t memory_limit_;
   size_t max_pool_size_;

mlx/backend/cuda/arange.cu

@@ -1,55 +0,0 @@
-// Copyright © 2025 Apple Inc.
-
-#include "mlx/backend/cuda/device.h"
-#include "mlx/backend/cuda/device/fp16_math.cuh"
-#include "mlx/backend/cuda/kernel_utils.cuh"
-#include "mlx/dtype_utils.h"
-#include "mlx/primitives.h"
-
-#include <nvtx3/nvtx3.hpp>
-#include <thrust/device_ptr.h>
-#include <thrust/transform.h>
-
-namespace mlx::core {
-
-namespace cu {
-
-template <typename T>
-struct Arange {
-  const T start;
-  const T step;
-
-  __device__ T operator()(uint32_t i) const {
-    return start + i * step;
-  }
-};
-
-} // namespace cu
-
-void Arange::eval_gpu(const std::vector<array>& inputs, array& out) {
-  nvtx3::scoped_range r("Arange::eval_gpu");
-  if (out.size() == 0) {
-    return;
-  }
-  out.set_data(allocator::malloc(out.nbytes()));
-
-  auto& encoder = cu::get_command_encoder(stream());
-  encoder.set_output_array(out);
-
-  auto capture = encoder.capture_context();
-  dispatch_int_float_types(out.dtype(), "Arange", [&](auto type_tag) {
-    using CTYPE = MLX_GET_TYPE(type_tag);
-    using OutType = cuda_type_t<CTYPE>;
-    CTYPE step =
-        static_cast<CTYPE>(start_ + step_) - static_cast<CTYPE>(start_);
-    thrust::transform(
-        cu::thrust_policy(encoder.stream()),
-        thrust::counting_iterator<uint32_t>(0),
-        thrust::counting_iterator<uint32_t>(out.data_size()),
-        thrust::device_pointer_cast(out.data<OutType>()),
-        cu::Arange<OutType>{
-            static_cast<OutType>(start_), static_cast<OutType>(step)});
-  });
-}
-
-} // namespace mlx::core

mlx/backend/cuda/arg_reduce.cu

@@ -1,8 +1,8 @@
 // Copyright © 2025 Apple Inc.
-
 #include "mlx/backend/common/utils.h"
 #include "mlx/backend/cuda/device.h"
 #include "mlx/backend/cuda/device/fp16_math.cuh"
+#include "mlx/backend/cuda/iterators/strided_iterator.cuh"
 #include "mlx/backend/cuda/kernel_utils.cuh"
 #include "mlx/dtype_utils.h"
 #include "mlx/primitives.h"
@@ -44,11 +44,8 @@ struct ArgMin {
   }
 
   template <int N>
-  __device__ IndexValPair<T> reduce_many(
-      IndexValPair<T> best,
-      const AlignedVector<T, N>& vals,
-      uint32_t offset) {
-#pragma unroll
+  __device__ IndexValPair<T>
+  reduce_many(IndexValPair<T> best, T (&vals)[N], uint32_t offset) {
     for (int i = 0; i < N; i++) {
       if (vals[i] < best.val) {
         best.val = vals[i];
@@ -77,11 +74,8 @@ struct ArgMax {
   }
 
   template <int N>
-  __device__ IndexValPair<T> reduce_many(
-      IndexValPair<T> best,
-      const AlignedVector<T, N>& vals,
-      uint32_t offset) {
-#pragma unroll
+  __device__ IndexValPair<T>
+  reduce_many(IndexValPair<T> best, T (&vals)[N], uint32_t offset) {
     for (int i = 0; i < N; i++) {
       if (vals[i] > best.val) {
         best.val = vals[i];
@@ -112,15 +106,16 @@ __global__ void arg_reduce_general(
 
   int64_t in_idx = elem_to_loc(index, shape.data(), in_strides.data(), ndim);
   int64_t out_idx = elem_to_loc(index, shape.data(), out_strides.data(), ndim);
-  in += in_idx;
 
   Op op;
   T init = op.init();
   IndexValPair<T> best{0, init};
 
   for (int r = 0; r < cuda::ceil_div(axis_size, BLOCK_DIM * N_READS); ++r) {
+    T vals[N_READS];
     auto tid = r * BLOCK_DIM + block.thread_index().x;
-    auto vals = load_vector<N_READS>(in, tid, axis_size, axis_stride, init);
+    cub::LoadDirectBlocked(
+        tid, strided_iterator(in + in_idx, axis_stride), vals, axis_size, init);
     best = op.reduce_many(best, vals, tid * N_READS);
   }
 

mlx/backend/cuda/binary.cu

@@ -28,7 +28,7 @@ __global__ void binary_ss(const In* a, const In* b, Out* out, IdxT size) {
     AlignedVector<Out, N_READS> out_vec;
 #pragma unroll
     for (int i = 0; i < N_READS; ++i) {
-      out_vec[i] = Op{}(a[0], b[0]);
+      out_vec.val[i] = Op{}(a[0], b[0]);
     }
 
     store_vector<N_READS>(out, index, out_vec);
@@ -49,7 +49,7 @@ __global__ void binary_sv(const In* a, const In* b, Out* out, IdxT size) {
     AlignedVector<Out, N_READS> out_vec;
 #pragma unroll
     for (int i = 0; i < N_READS; ++i) {
-      out_vec[i] = Op{}(a[0], b_vec[i]);
+      out_vec.val[i] = Op{}(a[0], b_vec.val[i]);
     }
 
     store_vector<N_READS>(out, index, out_vec);
@@ -70,7 +70,7 @@ __global__ void binary_vs(const In* a, const In* b, Out* out, IdxT size) {
     AlignedVector<Out, N_READS> out_vec;
 #pragma unroll
     for (int i = 0; i < N_READS; ++i) {
-      out_vec[i] = Op{}(a_vec[i], b[0]);
+      out_vec.val[i] = Op{}(a_vec.val[i], b[0]);
     }
 
     store_vector<N_READS>(out, index, out_vec);
@@ -92,7 +92,7 @@ __global__ void binary_vv(const In* a, const In* b, Out* out, IdxT size) {
     AlignedVector<Out, N_READS> out_vec;
 #pragma unroll
     for (int i = 0; i < N_READS; ++i) {
-      out_vec[i] = Op{}(a_vec[i], b_vec[i]);
+      out_vec.val[i] = Op{}(a_vec.val[i], b_vec.val[i]);
     }
 
     store_vector<N_READS>(out, index, out_vec);
@@ -128,7 +128,7 @@ __global__ void binary_g(
     int ndim) {
   IdxT index = cg::this_grid().thread_rank();
   if (index < size) {
-    auto [a_idx, b_idx] = elem_to_loc(
+    auto [a_idx, b_idx] = elem_to_loc_4d(
         index, shape.data(), a_strides.data(), b_strides.data(), ndim);
     out[index] = Op{}(a[a_idx], b[b_idx]);
   }
@@ -211,15 +211,12 @@ void binary_op_gpu_inplace(
                 int ndim = shape.size();
                 if (ndim <= 3) {
                   dispatch_1_2_3(ndim, [&](auto dims_constant) {
+                    auto kernel = cu::
+                        binary_g_nd<Op, InType, OutType, IdxT, dims_constant()>;
                     auto [num_blocks, block_dims] =
-                        get_launch_args(out, large());
+                        get_launch_args(kernel, out, large());
                     encoder.add_kernel_node(
-                        cu::binary_g_nd<
-                            Op,
-                            InType,
-                            OutType,
-                            IdxT,
-                            dims_constant()>,
+                        kernel,
                         num_blocks,
                         block_dims,
                         0,
@@ -232,9 +229,11 @@ void binary_op_gpu_inplace(
                         const_param<dims_constant()>(b_strides));
                   });
                 } else {
-                  auto [num_blocks, block_dims] = get_launch_args(out, large());
+                  auto kernel = cu::binary_g<Op, InType, OutType, IdxT>;
+                  auto [num_blocks, block_dims] =
+                      get_launch_args(kernel, out, large());
                   encoder.add_kernel_node(
-                      cu::binary_g<Op, InType, OutType, IdxT>,
+                      kernel,
                       num_blocks,
                       block_dims,
                       0,
@@ -251,7 +250,8 @@ void binary_op_gpu_inplace(
         } else {
           dispatch_bool(out.data_size() > UINT32_MAX, [&](auto large) {
             using IdxT = std::conditional_t<large(), int64_t, uint32_t>;
-            constexpr int N_READS = 16 / sizeof(InType);
+            // TODO: Choose optimized value based on type size.
+            constexpr int N_READS = 4;
             auto kernel = cu::binary_ss<Op, InType, OutType, IdxT, N_READS>;
             if (bopt == BinaryOpType::ScalarVector) {
               kernel = cu::binary_sv<Op, InType, OutType, IdxT, N_READS>;
@@ -261,7 +261,12 @@ void binary_op_gpu_inplace(
               kernel = cu::binary_vv<Op, InType, OutType, IdxT, N_READS>;
             }
             auto [num_blocks, block_dims] = get_launch_args(
-                out.data_size(), out.shape(), out.strides(), large(), N_READS);
+                kernel,
+                out.data_size(),
+                out.shape(),
+                out.strides(),
+                large(),
+                N_READS);
             encoder.add_kernel_node(
                 kernel,
                 num_blocks,

mlx/backend/cuda/binary_two.cu

@@ -33,8 +33,8 @@ binary_two_ss(const In* a, const In* b, Out* out_a, Out* out_b, IdxT size) {
 #pragma unroll
     for (int i = 0; i < N_READS; ++i) {
       auto out = Op{}(a[0], b[0]);
-      out_a_vec[i] = out[0];
-      out_b_vec[i] = out[1];
+      out_a_vec.val[i] = out[0];
+      out_b_vec.val[i] = out[1];
     }
 
     store_vector<N_READS>(out_a, index, out_a_vec);
@@ -60,9 +60,9 @@ binary_two_sv(const In* a, const In* b, Out* out_a, Out* out_b, IdxT size) {
     AlignedVector<Out, N_READS> out_b_vec;
 #pragma unroll
     for (int i = 0; i < N_READS; ++i) {
-      auto out = Op{}(a[0], b_vec[i]);
-      out_a_vec[i] = out[0];
-      out_b_vec[i] = out[1];
+      auto out = Op{}(a[0], b_vec.val[i]);
+      out_a_vec.val[i] = out[0];
+      out_b_vec.val[i] = out[1];
     }
 
     store_vector<N_READS>(out_a, index, out_a_vec);
@@ -88,9 +88,9 @@ binary_two_vs(const In* a, const In* b, Out* out_a, Out* out_b, IdxT size) {
     AlignedVector<Out, N_READS> out_b_vec;
 #pragma unroll
     for (int i = 0; i < N_READS; ++i) {
-      auto out = Op{}(a_vec[i], b[0]);
-      out_a_vec[i] = out[0];
-      out_b_vec[i] = out[1];
+      auto out = Op{}(a_vec.val[i], b[0]);
+      out_a_vec.val[i] = out[0];
+      out_b_vec.val[i] = out[1];
     }
 
     store_vector<N_READS>(out_a, index, out_a_vec);
@@ -117,9 +117,9 @@ binary_two_vv(const In* a, const In* b, Out* out_a, Out* out_b, IdxT size) {
     AlignedVector<Out, N_READS> out_b_vec;
 #pragma unroll
     for (int i = 0; i < N_READS; ++i) {
-      auto out = Op{}(a_vec[i], b_vec[i]);
-      out_a_vec[i] = out[0];
-      out_b_vec[i] = out[1];
+      auto out = Op{}(a_vec.val[i], b_vec.val[i]);
+      out_a_vec.val[i] = out[0];
+      out_b_vec.val[i] = out[1];
     }
 
     store_vector<N_READS>(out_a, index, out_a_vec);
@@ -160,7 +160,7 @@ __global__ void binary_two_g(
     int ndim) {
   IdxT index = cg::this_grid().thread_rank();
   if (index < size) {
-    auto [a_idx, b_idx] = elem_to_loc(
+    auto [a_idx, b_idx] = elem_to_loc_4d(
         index, shape.data(), a_strides.data(), b_strides.data(), ndim);
     auto out = Op{}(a[a_idx], b[b_idx]);
     out_a[index] = out[0];
@@ -227,15 +227,16 @@ void binary_two_op_gpu_inplace(
                 int ndim = shape.size();
                 if (ndim <= 3) {
                   dispatch_1_2_3(ndim, [&](auto dims_constant) {
+                    auto kernel = cu::binary_two_g_nd<
+                        Op,
+                        InType,
+                        OutType,
+                        IdxT,
+                        dims_constant()>;
                     auto [num_blocks, block_dims] =
-                        get_launch_args(out_a, large());
+                        get_launch_args(kernel, out_a, large());
                     encoder.add_kernel_node(
-                        cu::binary_two_g_nd<
-                            Op,
-                            InType,
-                            OutType,
-                            IdxT,
-                            dims_constant()>,
+                        kernel,
                         num_blocks,
                         block_dims,
                         0,
@@ -249,10 +250,11 @@ void binary_two_op_gpu_inplace(
                         const_param<dims_constant()>(b_strides));
                   });
                 } else {
+                  auto kernel = cu::binary_two_g<Op, InType, OutType, IdxT>;
                   auto [num_blocks, block_dims] =
-                      get_launch_args(out_a, large());
+                      get_launch_args(kernel, out_a, large());
                   encoder.add_kernel_node(
-                      cu::binary_two_g<Op, InType, OutType, IdxT>,
+                      kernel,
                       num_blocks,
                       block_dims,
                       0,
@@ -270,7 +272,8 @@ void binary_two_op_gpu_inplace(
         } else {
           dispatch_bool(out_a.data_size() > UINT32_MAX, [&](auto large) {
             using IdxT = std::conditional_t<large(), int64_t, uint32_t>;
-            constexpr int N_READS = 16 / sizeof(InType);
+            // TODO: Choose optimized value based on type size.
+            constexpr int N_READS = 4;
             auto kernel = cu::binary_two_ss<Op, InType, OutType, IdxT, N_READS>;
             if (bopt == BinaryOpType::ScalarVector) {
               kernel = cu::binary_two_sv<Op, InType, OutType, IdxT, N_READS>;
@@ -280,6 +283,7 @@ void binary_two_op_gpu_inplace(
               kernel = cu::binary_two_vv<Op, InType, OutType, IdxT, N_READS>;
             }
             auto [num_blocks, block_dims] = get_launch_args(
+                kernel,
                 out_a.data_size(),
                 out_a.shape(),
                 out_a.strides(),

mlx/backend/cuda/compiled.cpp

@@ -104,41 +104,10 @@ struct FusedKernelBuilder {
           "  }\n";
     }
 
-    // Vectorized read loop
-    if (contiguous) {
-      for (size_t i = 0; i < inputs.size(); ++i) {
-        const auto& x = inputs[i];
-        if (is_scalar(x) || is_constant(i)) {
-          continue;
-        }
-        const std::string& xname = namer.get_name(x);
-        std::string type = dtype_to_cuda_type(x.dtype());
-        os += fmt::format(
-            "  auto vec_{0} = load_vector<work_per_thread, {1}>({0} + index, 0, size - index, 0);\n",
-            xname,
-            type);
-      }
-    }
-
-    // Create some space for the outputs
-    for (const auto& x : outputs) {
-      const std::string& xname = namer.get_name(x);
-      std::string type = dtype_to_cuda_type(x.dtype());
-      os += fmt::format(
-          "  AlignedVector<{}, work_per_thread> vec_{};\n", type, xname);
-    }
-
     // Work loop
-    if (!contiguous) {
-      os +=
-          "\n"
-          "  for (int i = 0; i < work_per_thread && index < size; i++) {\n";
-    } else {
-      os +=
-          "\n"
-          "  #pragma unroll\n"
-          "  for (int i = 0; i < work_per_thread; i++) {\n";
-    }
+    os +=
+        "\n"
+        "  for (int i = 0; i < work_per_thread && index < size; i++) {\n";
 
     // Read inputs.
     for (size_t i = 0; i < inputs.size(); ++i) {
@@ -153,7 +122,7 @@ struct FusedKernelBuilder {
       } else if (is_scalar(x)) {
         value = fmt::format("{}[0]", xname);
       } else if (contiguous) {
-        value = fmt::format("vec_{}[i]", xname);
+        value = fmt::format("{}[index]", xname);
       } else {
         value = fmt::format("{}[{}_idx]", xname, xname);
       }
@@ -181,30 +150,25 @@ struct FusedKernelBuilder {
 
     // Write output.
     for (const auto& x : outputs) {
-      os += fmt::format("    vec_{0}[i] = tmp_{0};\n", namer.get_name(x));
+      os += fmt::format("    {0}[index] = tmp_{0};\n", namer.get_name(x));
     }
 
     // End of work loop
+    os +=
+        "\n"
+        "    index++;\n";
     if (!contiguous) {
-      os += "\n";
       for (size_t i = 0; i < inputs.size(); ++i) {
         const auto& x = inputs[i];
         const std::string& xname = namer.get_name(x);
         if (is_scalar(x) || is_constant(i)) {
           continue;
         }
-        os += fmt::format("    {0}_idx += {0}_strides[NDIM - 1];\n", xname);
+        os += "    " + xname + "_idx += " + xname + "_strides[NDIM - 1];\n";
       }
     }
     os += "  }\n";
 
-    // Store the output to global memory
-    for (const auto& x : outputs) {
-      os += fmt::format(
-          "  store_vector({0} + index, 0, vec_{0}, size - index);\n",
-          namer.get_name(x));
-    }
-
     os += "}\n";
   }
 };
@@ -228,15 +192,6 @@ void Compiled::eval_gpu(
   nvtx3::scoped_range r("Compiled::eval_gpu");
   auto& s = stream();
 
-  // Determine the work per thread for the vectorized reads/writes. We take it
-  // as 16 over the max itemsize for the outputs. Another heuristic could be
-  // over the max itemsize of all arrays.
-  int max_size = 1;
-  for (const auto& x : outputs) {
-    max_size = (max_size > x.itemsize()) ? max_size : x.itemsize();
-  }
-  int work_per_thread = 16 / max_size;
-
   cu::JitModule& mod = cu::get_jit_module(s.device, lib_name(), [&]() {
     // Build source code.
     cu::FusedKernelBuilder builder{
@@ -250,23 +205,28 @@ void Compiled::eval_gpu(
     builder.os += "\n} // namespace mlx::core::cu\n";
     // Build kernel names.
     std::vector<std::string> kernel_names;
-    kernel_names.push_back(fmt::format(
-        "mlx::core::cu::{}_contiguous<uint32_t, {}>",
-        lib_name(),
-        work_per_thread));
-    kernel_names.push_back(fmt::format(
-        "mlx::core::cu::{}_contiguous<int64_t, {}>",
-        lib_name(),
-        work_per_thread));
-    for (auto wpt : std::array<int, 2>{1, work_per_thread}) {
+    for (auto work_per_thread : std::array<int, 2>{1, 4}) {
+      kernel_names.push_back(fmt::format(
+          "mlx::core::cu::{}_contiguous<uint32_t, {}>",
+          lib_name(),
+          work_per_thread));
+      kernel_names.push_back(fmt::format(
+          "mlx::core::cu::{}_contiguous<int64_t, {}>",
+          lib_name(),
+          work_per_thread));
       for (int i = 1; i <= MAX_NDIM; ++i) {
         kernel_names.push_back(fmt::format(
-            "mlx::core::cu::{}_strided<{}, uint32_t, {}>", lib_name(), i, wpt));
+            "mlx::core::cu::{}_strided<{}, uint32_t, {}>",
+            lib_name(),
+            i,
+            work_per_thread));
         kernel_names.push_back(fmt::format(
-            "mlx::core::cu::{}_strided<{}, int64_t, {}>", lib_name(), i, wpt));
+            "mlx::core::cu::{}_strided<{}, int64_t, {}>",
+            lib_name(),
+            i,
+            work_per_thread));
       }
     }
-
     return std::make_pair(std::move(builder.os), std::move(kernel_names));
   });
 
@@ -309,6 +269,7 @@ void Compiled::eval_gpu(
   }
 
   // Choose work per thread
+  int work_per_thread = 4;
   if (!contiguous && shape.back() % work_per_thread != 0) {
     work_per_thread = 1;
   }
@@ -333,7 +294,7 @@ void Compiled::eval_gpu(
 
   auto kernel = mod.get_kernel(kernel_name);
   auto [num_blocks, block_dims] =
-      get_launch_args(outputs[0], large, work_per_thread);
+      get_launch_args(kernel, outputs[0], large, work_per_thread);
   encoder.add_kernel_node(kernel, num_blocks, block_dims, 0, args.args());
 }
 

mlx/backend/cuda/conv.cpp

@@ -1,546 +0,0 @@
-// Copyright © 2025 Apple Inc.
-
-#include "mlx/backend/cuda/device.h"
-#include "mlx/backend/cuda/device/config.h"
-#include "mlx/backend/cuda/lru_cache.h"
-#include "mlx/backend/gpu/copy.h"
-#include "mlx/dtype_utils.h"
-#include "mlx/primitives.h"
-
-// cudnn_frontend.h redefines this macro.
-#undef CHECK_CUDA_ERROR
-
-#include <cudnn_frontend.h>
-#include <cudnn_frontend_find_plan.h>
-#include <fmt/format.h>
-#include <nvtx3/nvtx3.hpp>
-
-#include <cassert>
-
-namespace mlx::core {
-
-namespace {
-
-// Not all engines support it so can not use this API now.
-#define MLX_USE_CUDNN_NATIVE_CUDA_GRAPH_API 0
-
-// Alias for better readability.
-#define CONV_FORWARD CUDNN_BACKEND_OPERATION_CONVOLUTION_FORWARD_DESCRIPTOR
-#define CONV_BACKWARD_INPUT \
-  CUDNN_BACKEND_OPERATION_CONVOLUTION_BACKWARD_DATA_DESCRIPTOR
-#define CONV_BACKWARD_WEIGHT \
-  CUDNN_BACKEND_OPERATION_CONVOLUTION_BACKWARD_FILTER_DESCRIPTOR
-
-struct ConvCacheKey {
-  int device_id;
-  cudnnDataType_t cudnn_dtype;
-  std::array<int, MAX_NDIM> input_shape;
-  std::array<int, MAX_NDIM> weight_shape;
-  std::array<int, MAX_NDIM> stride;
-  std::array<int, MAX_NDIM> padding_lo;
-  std::array<int, MAX_NDIM> padding_hi;
-  std::array<int, MAX_NDIM> dilation;
-  int groups;
-  bool flip;
-  uint8_t input_alignment;
-  uint8_t weight_alignment;
-  uint8_t output_alignment;
-};
-
-auto& conv_cache() {
-  static LRUBytesKeyCache<
-      ConvCacheKey,
-      std::pair<cudnnBackendDescriptorType_t, cudnn_frontend::ExecutionPlan>>
-      cache(/* capacity */ 128);
-  return cache;
-}
-
-template <typename T, typename Vec>
-inline SmallVector<T> convert_vector(const Vec& vec) {
-  return SmallVector<T>(vec.begin(), vec.end());
-}
-
-template <typename T, template <typename U> class Vec>
-inline std::array<T, MAX_NDIM> fixed_vector(const Vec<T>& vec) {
-  if (vec.size() > MAX_NDIM) {
-    throw std::runtime_error(
-        fmt::format("ndim can not be larger than {}.", MAX_NDIM));
-  }
-  std::array<T, MAX_NDIM> result = {};
-  std::copy_n(vec.begin(), vec.size(), result.begin());
-  return result;
-}
-
-auto nhwc_to_nchw(const array& x) {
-  auto shape = convert_vector<int64_t>(x.shape());
-  shape.insert(shape.begin() + 1, shape.back());
-  shape.erase(shape.end() - 1);
-  auto strides = convert_vector<int64_t>(x.strides());
-  strides.insert(strides.begin() + 1, strides.back());
-  strides.erase(strides.end() - 1);
-  return std::make_tuple(std::move(shape), std::move(strides));
-}
-
-inline cudnnDataType_t dtype_to_cudnn_type(Dtype dtype) {
-  switch (dtype) {
-    case int8:
-      return CUDNN_DATA_INT8;
-    case int32:
-      return CUDNN_DATA_INT32;
-    case uint8:
-      return CUDNN_DATA_UINT8;
-    case float16:
-      return CUDNN_DATA_HALF;
-    case bfloat16:
-      return CUDNN_DATA_BFLOAT16;
-    case float32:
-      return CUDNN_DATA_FLOAT;
-    case float64:
-      return CUDNN_DATA_DOUBLE;
-    default:
-      throw std::runtime_error(fmt::format(
-          "Unsupported dtype in Convolution: {}.", dtype_to_string(dtype)));
-  }
-}
-
-inline uint8_t get_alignment(const array& x) {
-  uint8_t alignment = 1;
-  uintptr_t address = reinterpret_cast<uintptr_t>(x.data<void>());
-  for (; alignment < 32; alignment *= 2) {
-    if (address % (alignment * 2)) {
-      return alignment;
-    }
-  }
-  return alignment;
-}
-
-inline cudnn_frontend::Tensor build_tensor(int64_t id, const array& x) {
-  auto [shape, strides] = nhwc_to_nchw(x);
-  return cudnn_frontend::TensorBuilder()
-      .setDim(shape.size(), shape.data())
-      .setStrides(strides.size(), strides.data())
-      .setId(id)
-      .setAlignment(get_alignment(x))
-      .setDataType(dtype_to_cudnn_type(x.dtype()))
-      .build();
-}
-
-cudnn_frontend::EngineConfigList get_engine_configs(
-    cudnnBackendDescriptorType_t backend_type,
-    Dtype dtype,
-    cudnn_frontend::OperationGraph& op_graph,
-    bool use_fallback = false) {
-  cudnn_frontend::GeneratorSource source;
-  if (use_fallback) {
-    source = [&backend_type](cudnn_frontend::OperationGraph& op_graph) {
-      auto fallback = cudnn_frontend::EngineFallbackListBuilder()
-                          .setOperationGraph(op_graph)
-                          .setOperation(backend_type)
-                          .build();
-      return fallback.getFallbackList();
-    };
-  } else {
-    source = [](cudnn_frontend::OperationGraph& op_graph) {
-      auto heuristics = cudnn_frontend::EngineHeuristicsBuilder()
-                            .setOperationGraph(op_graph)
-                            .setHeurMode(CUDNN_HEUR_MODE_A)
-                            .build();
-      return heuristics.getEngineConfig(heuristics.getEngineConfigCount());
-    };
-  }
-
-  cudnn_frontend::EngineConfigGenerator generator(1, &source);
-  auto configs = generator.generate_engine_config(op_graph);
-
-  cudnn_frontend::EngineConfigList filtered_configs;
-  cudnn_frontend::filter(configs, filtered_configs, [dtype](auto c) {
-    if (cudnn_frontend::hasNumericalNote<
-            CUDNN_NUMERICAL_NOTE_DOWN_CONVERT_INPUTS>(c)) {
-      return true;
-    }
-    if (cudnn_frontend::hasNumericalNote<CUDNN_NUMERICAL_NOTE_TENSOR_CORE>(c) &&
-        dtype == float32 && !env::enable_tf32()) {
-      return true;
-    }
-    return false;
-  });
-  return filtered_configs;
-}
-
-bool execute_plan(
-    cu::CommandEncoder& encoder,
-    cudnn_frontend::ExecutionPlan& plan,
-    array& x,
-    array& w,
-    array& y) {
-  int workspace_size = plan.getWorkspaceSize();
-  array workspace(allocator::malloc(workspace_size), {workspace_size}, uint8);
-
-  int64_t uids[3] = {'x', 'w', 'y'};
-  void* data_ptrs[3] = {
-      x.data<void>(),
-      w.data<void>(),
-      y.data<void>(),
-  };
-
-  auto variantPack = cudnn_frontend::VariantPackBuilder()
-                         .setWorkspacePointer(workspace.data<void>())
-                         .setDataPointers(3, data_ptrs)
-                         .setUids(3, uids)
-                         .build();
-
-  auto handle = encoder.device().cudnn_handle();
-  cudnnSetStream(handle, encoder.stream());
-
-#if CUDNN_VERSION >= 90500 && MLX_USE_CUDNN_NATIVE_CUDA_GRAPH_API
-  cudaGraph_t graph;
-  cudaGraphCreate(&graph, 0);
-  std::unique_ptr<cudaGraph_t, void (*)(cudaGraph_t*)> graph_freer(
-      &graph, [](cudaGraph_t* p) { cudaGraphDestroy(*p); });
-  if (cudnnBackendPopulateCudaGraph(
-          handle, plan.get_raw_desc(), variantPack.get_raw_desc(), graph) !=
-      CUDNN_STATUS_SUCCESS) {
-    return false;
-  }
-  encoder.add_graph_node(graph);
-#else
-  auto capture = encoder.capture_context();
-  if (cudnnBackendExecute(
-          handle, plan.get_raw_desc(), variantPack.get_raw_desc()) !=
-      CUDNN_STATUS_SUCCESS) {
-    // Discard the captured graph when failed.
-    capture.discard = true;
-    return false;
-  }
-#endif
-
-  encoder.add_temporary(workspace);
-  return true;
-}
-
-bool try_engines(
-    cu::CommandEncoder& encoder,
-    const ConvCacheKey& cache_key,
-    cudnnBackendDescriptorType_t backend_type,
-    cudnn_frontend::EngineConfigList& configs,
-    const std::string& op_graph_tag,
-    array& x,
-    array& w,
-    array& y) {
-  for (auto& config : configs) {
-    try {
-      auto plan = cudnn_frontend::ExecutionPlanBuilder()
-                      .setHandle(encoder.device().cudnn_handle())
-                      .setEngineConfig(config, op_graph_tag)
-                      .build();
-      if (execute_plan(encoder, plan, x, w, y)) {
-        conv_cache().emplace(
-            cache_key, std::make_pair(backend_type, std::move(plan)));
-        return true;
-      }
-    } catch (cudnn_frontend::cudnnException& error) {
-      if (error.getCudnnStatus() != CUDNN_STATUS_NOT_SUPPORTED) {
-        throw;
-      }
-    }
-  }
-  return false;
-}
-
-auto get_conv_op_settings(
-    cudnnBackendDescriptorType_t backend_type,
-    array& x,
-    array& w,
-    array& y,
-    const std::vector<int>& kernel_strides,
-    const std::vector<int>& padding_lo_,
-    const std::vector<int>& padding_hi_,
-    const std::vector<int>& kernel_dilation,
-    const std::vector<int>& input_dilation) {
-  auto padding_lo = convert_vector<int64_t>(padding_lo_);
-  auto padding_hi = convert_vector<int64_t>(padding_hi_);
-
-  if (backend_type == CONV_BACKWARD_INPUT) {
-    for (int i = 0; i < padding_lo.size(); ++i) {
-      int wt_size = 1 + kernel_dilation[i] * (w.shape(1 + i) - 1);
-      padding_lo[i] = wt_size - padding_lo[i] - 1;
-      int in_size = 1 + kernel_strides[i] * (x.shape(1 + i) - 1);
-      int out_size = 1 + input_dilation[i] * (y.shape(1 + i) - 1);
-      padding_hi[i] = out_size - in_size + padding_hi[i];
-    }
-    return std::make_tuple(
-        convert_vector<int64_t>(input_dilation),
-        std::move(padding_lo),
-        std::move(padding_hi),
-        convert_vector<int64_t>(kernel_dilation));
-
-  } else if (backend_type == CONV_BACKWARD_WEIGHT) {
-    padding_hi = padding_lo;
-    return std::make_tuple(
-        convert_vector<int64_t>(kernel_dilation),
-        std::move(padding_lo),
-        std::move(padding_hi),
-        convert_vector<int64_t>(kernel_strides));
-
-  } else {
-    return std::make_tuple(
-        convert_vector<int64_t>(kernel_strides),
-        std::move(padding_lo),
-        std::move(padding_hi),
-        convert_vector<int64_t>(kernel_dilation));
-  }
-}
-
-std::optional<cudnn_frontend::OperationGraph> build_op_graph(
-    cu::CommandEncoder& encoder,
-    cudnnBackendDescriptorType_t backend_type,
-    Dtype dtype,
-    array& x,
-    array& w,
-    array& y,
-    const SmallVector<int64_t>& stride,
-    const SmallVector<int64_t>& padding_lo,
-    const SmallVector<int64_t>& padding_hi,
-    const SmallVector<int64_t>& dilation) {
-  try {
-    auto compute_dtype = (dtype == float16 || dtype == bfloat16)
-        ? CUDNN_DATA_FLOAT
-        : dtype_to_cudnn_type(dtype);
-    auto conv_desc = cudnn_frontend::ConvDescBuilder()
-                         .setDataType(compute_dtype)
-                         .setMathMode(CUDNN_CROSS_CORRELATION)
-                         .setNDims(stride.size())
-                         .setStrides(stride.size(), stride.data())
-                         .setPrePadding(padding_lo.size(), padding_lo.data())
-                         .setPostPadding(padding_hi.size(), padding_hi.data())
-                         .setDilation(dilation.size(), dilation.data())
-                         .build();
-
-    auto op = cudnn_frontend::OperationBuilder(backend_type)
-                  .setxDesc(build_tensor('x', x))
-                  .setwDesc(build_tensor('w', w))
-                  .setyDesc(build_tensor('y', y))
-                  .setcDesc(conv_desc)
-                  .build();
-
-    std::array<cudnn_frontend::Operation const*, 1> ops = {&op};
-    return cudnn_frontend::OperationGraphBuilder()
-        .setHandle(encoder.device().cudnn_handle())
-        .setOperationGraph(ops.size(), ops.data())
-        .build();
-  } catch (cudnn_frontend::cudnnException& error) {
-    if (error.getCudnnStatus() != CUDNN_STATUS_BAD_PARAM) {
-      throw;
-    }
-    return std::nullopt;
-  }
-}
-
-// Do necessary transposes and copies to prepare the inputs and outputs for
-// building the cuDNN conv op. It is safe to be called multiple times in one
-// eval_gpu, with cost of possible redundant copies.
-std::tuple<array, array, array> prepare_args(
-    cu::CommandEncoder& encoder,
-    cudnnBackendDescriptorType_t backend_type,
-    array in,
-    array wt,
-    array out,
-    Stream s) {
-  // Transpose the args depending on the backend type.
-  // TODO: Handle groups.
-  if (backend_type == CONV_BACKWARD_INPUT) {
-    wt = swapaxes_in_eval(wt, 0, -1);
-  } else if (backend_type == CONV_BACKWARD_WEIGHT) {
-    in = swapaxes_in_eval(in, 0, -1);
-    wt = swapaxes_in_eval(wt, 0, -1);
-    // Create a contiguous array that shares the data with |out|, but with dim
-    // C_in and C_out swapped.
-    Shape shape(out.shape());
-    std::swap(shape.front(), shape.back());
-    Strides strides(shape.size(), 1);
-    for (int i = shape.size() - 2; i >= 0; --i) {
-      strides[i] = shape[i + 1] * strides[i + 1];
-    }
-    array intermediate(std::move(shape), out.dtype(), nullptr, {});
-    intermediate.copy_shared_buffer(
-        out, std::move(strides), {true, true, false}, out.data_size());
-    out = intermediate;
-  }
-
-  // cuDNN requires contiguous input.
-  if (!in.flags().row_contiguous) {
-    in = contiguous_copy_gpu(in, s);
-    encoder.add_temporary(in);
-  }
-  if (!wt.flags().row_contiguous) {
-    wt = contiguous_copy_gpu(wt, s);
-    encoder.add_temporary(wt);
-  }
-
-  return {std::move(in), std::move(wt), std::move(out)};
-}
-
-// Get the x/w/y args from the in/wt/out args depending on backend type.
-inline std::tuple<array&, array&, array&> dispatch_args(
-    cudnnBackendDescriptorType_t backend_type,
-    array& in,
-    array& wt,
-    array& out) {
-  switch (backend_type) {
-    case CONV_BACKWARD_INPUT:
-      return {out, wt, in};
-    case CONV_BACKWARD_WEIGHT:
-      return {in, out, wt};
-    default:
-      return {in, wt, out};
-  }
-}
-
-// Register inputs and outputs before actually running conv op. Can only be
-// called once per eval_gpu.
-void register_args(
-    cu::CommandEncoder& encoder,
-    cudnnBackendDescriptorType_t backend_type,
-    array& in,
-    array& wt,
-    array& intermediate_out,
-    array& final_out) {
-  encoder.set_input_array(in);
-  encoder.set_input_array(wt);
-  encoder.set_output_array(final_out);
-
-  if (backend_type == CONV_BACKWARD_WEIGHT) {
-    // Turn |out| into a strided array, which will have C_in and C_out swapped
-    // in vjp and the final |grad_weight| will then be contiguous.
-    Strides strides = intermediate_out.strides();
-    std::swap(strides.front(), strides.back());
-    final_out.copy_shared_buffer(
-        intermediate_out,
-        std::move(strides),
-        {false, false, false},
-        intermediate_out.data_size());
-  }
-}
-
-} // namespace
-
-void Convolution::eval_gpu(const std::vector<array>& inputs, array& out_) {
-  nvtx3::scoped_range r("Convolution::eval_gpu");
-  if (out_.size() == 0) {
-    return;
-  }
-
-  assert(inputs.size() == 2);
-  array in = inputs[0];
-  array wt = inputs[1];
-  array out = out_;
-  out.set_data(allocator::malloc(out.nbytes()));
-  Dtype dtype = out.dtype();
-
-  auto& s = stream();
-  auto& encoder = cu::get_command_encoder(s);
-
-  // Search cache.
-  ConvCacheKey cache_key{
-      encoder.device().cuda_device(),
-      dtype_to_cudnn_type(dtype),
-      fixed_vector(in.shape()),
-      fixed_vector(wt.shape()),
-      fixed_vector(kernel_strides_),
-      fixed_vector(padding_lo_),
-      fixed_vector(padding_hi_),
-      fixed_vector(kernel_dilation_),
-      groups_,
-      flip_,
-      get_alignment(in),
-      get_alignment(wt),
-      get_alignment(out)};
-  if (auto it = conv_cache().find(cache_key); it != conv_cache().end()) {
-    auto& [backend_type, plan] = it->second;
-    std::tie(in, wt, out) = prepare_args(encoder, backend_type, in, wt, out, s);
-    register_args(encoder, backend_type, in, wt, out, out_);
-    auto [x, w, y] = dispatch_args(backend_type, in, wt, out);
-    if (!execute_plan(encoder, plan, x, w, y)) {
-      throw std::runtime_error("[conv] Cached plan failed to execute.");
-    }
-    return;
-  }
-
-  // There is no reliable way to deduce the proper cuDNN backend for the
-  // convolution, so we make a best guess and then try.
-  SmallVector<cudnnBackendDescriptorType_t, 2> try_backends;
-  if (flip_) {
-    // When weight is flipped, we assume it is backward input convolution.
-    try_backends.push_back(CONV_BACKWARD_INPUT);
-  } else {
-    // Otherwise it could be backward weight convolution or forward convolution,
-    // mathematically there is no difference so we have to use heuristics.
-    // Empirically backward convolutions have large kernel dimensions, and
-    // usually have |in| and |wt| transposed.
-    if (!in.flags().row_contiguous && !wt.flags().row_contiguous &&
-        wt.shape(2) > out.shape(2)) {
-      try_backends = {CONV_BACKWARD_WEIGHT, CONV_FORWARD};
-    } else {
-      try_backends = {CONV_FORWARD, CONV_BACKWARD_WEIGHT};
-    }
-  }
-
-  // Try to build op graph.
-  cudnnBackendDescriptorType_t backend_type;
-  std::optional<cudnn_frontend::OperationGraph> op_graph;
-  for (auto try_backend : try_backends) {
-    auto [in_copy, wt_copy, out_copy] =
-        prepare_args(encoder, try_backend, in, wt, out, s);
-    auto [x, w, y] = dispatch_args(try_backend, in_copy, wt_copy, out_copy);
-    auto [stride, padding_lo, padding_hi, dilation] = get_conv_op_settings(
-        try_backend,
-        x,
-        w,
-        y,
-        kernel_strides_,
-        padding_lo_,
-        padding_hi_,
-        kernel_dilation_,
-        input_dilation_);
-    op_graph = build_op_graph(
-        encoder,
-        try_backend,
-        dtype,
-        x,
-        w,
-        y,
-        stride,
-        padding_lo,
-        padding_hi,
-        dilation);
-    if (op_graph) {
-      backend_type = try_backend;
-      in = std::move(in_copy);
-      wt = std::move(wt_copy);
-      out = std::move(out_copy);
-      break;
-    }
-  }
-  if (!op_graph) {
-    throw std::runtime_error("[conv] Can not build op graph.");
-  }
-
-  // Get ready to execute the graph.
-  register_args(encoder, backend_type, in, wt, out, out_);
-
-  // Try to run plans based on heuristics.
-  auto configs = get_engine_configs(backend_type, dtype, *op_graph);
-  auto tag = op_graph->getTag();
-  auto [x, w, y] = dispatch_args(backend_type, in, wt, out);
-  if (try_engines(encoder, cache_key, backend_type, configs, tag, x, w, y)) {
-    return;
-  }
-  // Then try fallback plans.
-  configs = get_engine_configs(backend_type, dtype, *op_graph);
-  if (try_engines(encoder, cache_key, backend_type, configs, tag, x, w, y)) {
-    return;
-  }
-  throw std::runtime_error("[conv] Unable to find a working engine.");
-}
-
-} // namespace mlx::core

mlx/backend/cuda/copy/copy_contiguous.cu

@@ -22,7 +22,7 @@ __global__ void copy_s(const In* in, Out* out, IdxT size) {
     AlignedVector<Out, N_READS> out_vec;
 #pragma unroll
     for (int i = 0; i < N_READS; ++i) {
-      out_vec[i] = cast_to<Out>(in[0]);
+      out_vec.val[i] = cast_to<Out>(in[0]);
     }
 
     store_vector<N_READS>(out, index, out_vec);
@@ -43,7 +43,7 @@ __global__ void copy_v(const In* in, Out* out, IdxT size) {
     AlignedVector<Out, N_READS> out_vec;
 #pragma unroll
     for (int i = 0; i < N_READS; ++i) {
-      out_vec[i] = cast_to<Out>(in_vec[i]);
+      out_vec.val[i] = cast_to<Out>(in_vec.val[i]);
     }
 
     store_vector<N_READS>(out, index, out_vec);
@@ -65,13 +65,19 @@ void copy_contiguous(
         using InType = cuda_type_t<MLX_GET_TYPE(in_type_tag)>;
         using OutType = cuda_type_t<MLX_GET_TYPE(out_type_tag)>;
         using IdxT = std::conditional_t<large(), int64_t, uint32_t>;
-        constexpr int N_READS = 16 / sizeof(InType);
+        // TODO: Choose optimized value based on type size.
+        constexpr int N_READS = 4;
         auto kernel = cu::copy_s<InType, OutType, IdxT, N_READS>;
         if (ctype == CopyType::Vector) {
           kernel = cu::copy_v<InType, OutType, IdxT, N_READS>;
         }
         auto [num_blocks, block_dims] = get_launch_args(
-            out.data_size(), out.shape(), out.strides(), large(), N_READS);
+            kernel,
+            out.data_size(),
+            out.shape(),
+            out.strides(),
+            large(),
+            N_READS);
         encoder.add_kernel_node(
             kernel,
             num_blocks,

mlx/backend/cuda/copy/copy_general.cu

@@ -37,7 +37,7 @@ __global__ void copy_gg(
     int ndim) {
   IdxT index = cg::this_grid().thread_rank();
   if (index < size) {
-    auto [idx_in, idx_out] = elem_to_loc(
+    auto [idx_in, idx_out] = elem_to_loc_4d(
         index, shape.data(), strides_in.data(), strides_out.data(), ndim);
     out[idx_out] = CastOp<In, Out>{}(in[idx_in]);
   }
@@ -71,10 +71,12 @@ void copy_general(
               data_size *= s;
             if (ndim <= 3) {
               dispatch_1_2_3(ndim, [&](auto ndim_constant) {
-                auto [num_blocks, block_dims] =
-                    get_launch_args(data_size, shape, out.strides(), large());
+                auto kernel =
+                    cu::copy_gg_nd<InType, OutType, IdxT, ndim_constant()>;
+                auto [num_blocks, block_dims] = get_launch_args(
+                    kernel, data_size, shape, out.strides(), large());
                 encoder.add_kernel_node(
-                    cu::copy_gg_nd<InType, OutType, IdxT, ndim_constant()>,
+                    kernel,
                     num_blocks,
                     block_dims,
                     0,
@@ -86,10 +88,11 @@ void copy_general(
                     const_param<ndim_constant()>(strides_out));
               });
             } else { // ndim >= 4
-              auto [num_blocks, block_dims] =
-                  get_launch_args(data_size, shape, out.strides(), large());
+              auto kernel = cu::copy_gg<InType, OutType, IdxT>;
+              auto [num_blocks, block_dims] = get_launch_args(
+                  kernel, data_size, shape, out.strides(), large());
               encoder.add_kernel_node(
-                  cu::copy_gg<InType, OutType, IdxT>,
+                  kernel,
                   num_blocks,
                   block_dims,
                   0,

mlx/backend/cuda/copy/copy_general_dynamic.cu

@@ -41,7 +41,7 @@ __global__ void copy_gg_dynamic(
     const int64_t* offset_out) {
   IdxT index = cg::this_grid().thread_rank();
   if (index < size) {
-    auto [idx_in, idx_out] = elem_to_loc(
+    auto [idx_in, idx_out] = elem_to_loc_4d(
         index, shape.data(), strides_in.data(), strides_out.data(), ndim);
     out[idx_out + *offset_out] = CastOp<In, Out>{}(in[idx_in + *offset_in]);
   }
@@ -74,13 +74,12 @@ void copy_general_dynamic(
             int ndim = shape.size();
             if (ndim <= 3) {
               dispatch_1_2_3(ndim, [&](auto dims_constant) {
-                auto [num_blocks, block_dims] = get_launch_args(out, large());
+                auto kernel = cu::
+                    copy_gg_dynamic_nd<InType, OutType, IdxT, dims_constant()>;
+                auto [num_blocks, block_dims] =
+                    get_launch_args(kernel, out, large());
                 encoder.add_kernel_node(
-                    cu::copy_gg_dynamic_nd<
-                        InType,
-                        OutType,
-                        IdxT,
-                        dims_constant()>,
+                    kernel,
                     num_blocks,
                     block_dims,
                     0,
@@ -94,9 +93,11 @@ void copy_general_dynamic(
                     dynamic_offset_out.data<int64_t>());
               });
             } else { // ndim >= 4
-              auto [num_blocks, block_dims] = get_launch_args(out, large());
+              auto kernel = cu::copy_gg_dynamic<InType, OutType, IdxT>;
+              auto [num_blocks, block_dims] =
+                  get_launch_args(kernel, out, large());
               encoder.add_kernel_node(
-                  cu::copy_gg_dynamic<InType, OutType, IdxT>,
+                  kernel,
                   num_blocks,
                   block_dims,
                   0,

mlx/backend/cuda/copy/copy_general_input.cu

@@ -34,7 +34,7 @@ __global__ void copy_g(
     int ndim) {
   IdxT index = cg::this_grid().thread_rank();
   if (index < size) {
-    IdxT idx_in = elem_to_loc(index, shape.data(), strides_in.data(), ndim);
+    IdxT idx_in = elem_to_loc_4d(index, shape.data(), strides_in.data(), ndim);
     out[index] = CastOp<In, Out>{}(in[idx_in]);
   }
 }
@@ -63,9 +63,12 @@ void copy_general_input(
             int ndim = shape.size();
             if (ndim <= 3) {
               dispatch_1_2_3(ndim, [&](auto dims_constant) {
-                auto [num_blocks, block_dims] = get_launch_args(out, large());
+                auto kernel =
+                    cu::copy_g_nd<InType, OutType, IdxT, dims_constant()>;
+                auto [num_blocks, block_dims] =
+                    get_launch_args(kernel, out, large());
                 encoder.add_kernel_node(
-                    cu::copy_g_nd<InType, OutType, IdxT, dims_constant()>,
+                    kernel,
                     num_blocks,
                     block_dims,
                     0,
@@ -76,9 +79,11 @@ void copy_general_input(
                     const_param<dims_constant()>(strides_in));
               });
             } else { // ndim >= 4
-              auto [num_blocks, block_dims] = get_launch_args(out, large());
+              auto kernel = cu::copy_g<InType, OutType, IdxT>;
+              auto [num_blocks, block_dims] =
+                  get_launch_args(kernel, out, large());
               encoder.add_kernel_node(
-                  cu::copy_g<InType, OutType, IdxT>,
+                  kernel,
                   num_blocks,
                   block_dims,
                   0,

mlx/backend/cuda/device.cpp

@@ -1,7 +1,6 @@
 // Copyright © 2025 Apple Inc.
 
 #include "mlx/backend/cuda/device.h"
-#include "mlx/backend/cuda/jit_module.h"
 #include "mlx/backend/cuda/worker.h"
 #include "mlx/utils.h"
 
@@ -10,23 +9,12 @@
 #include <future>
 #include <unordered_set>
 
-namespace mlx::core::cu {
-
-namespace {
+namespace mlx::core {
 
 // Can be tuned with MLX_MAX_OPS_PER_BUFFER
 // This should be less than 255
 constexpr int default_max_nodes_per_graph = 20;
 
-#define CHECK_CUDNN_ERROR(cmd) check_cudnn_error(#cmd, (cmd))
-
-void check_cudnn_error(const char* name, cudnnStatus_t err) {
-  if (err != CUDNN_STATUS_SUCCESS) {
-    throw std::runtime_error(
-        fmt::format("{} failed: {}.", name, cudnnGetErrorString(err)));
-  }
-}
-
 int cuda_graph_cache_size() {
   static int cache_size = []() {
     return env::get_var("MLX_CUDA_GRAPH_CACHE_SIZE", 100);
@@ -34,7 +22,7 @@ int cuda_graph_cache_size() {
   return cache_size;
 }
 
-} // namespace
+namespace cu {
 
 Device::Device(int device) : device_(device) {
   CHECK_CUDA_ERROR(cudaDeviceGetAttribute(
@@ -52,18 +40,11 @@ Device::Device(int device) : device_(device) {
   }
   // The cublasLt handle is used by matmul.
   make_current();
-  CHECK_CUBLAS_ERROR(cublasLtCreate(&lt_));
-  // The cudnn handle is used by Convolution.
-  CHECK_CUDNN_ERROR(cudnnCreate(&cudnn_));
-
-  // Initialize the jit module cache here ensures it is not
-  // unloaded before any evaluation is done
-  get_jit_module_cache();
+  cublasLtCreate(&lt_);
 }
 
 Device::~Device() {
-  CHECK_CUDNN_ERROR(cudnnDestroy(cudnn_));
-  CHECK_CUBLAS_ERROR(cublasLtDestroy(lt_));
+  cublasLtDestroy(lt_);
 }
 
 void Device::make_current() {
@@ -85,19 +66,29 @@ CommandEncoder& Device::get_command_encoder(Stream s) {
 }
 
 CommandEncoder::CaptureContext::CaptureContext(CommandEncoder& enc) : enc(enc) {
-  enc.device().make_current();
   CHECK_CUDA_ERROR(
       cudaStreamBeginCapture(enc.stream(), cudaStreamCaptureModeGlobal));
 }
 
 CommandEncoder::CaptureContext::~CaptureContext() {
   CHECK_CUDA_ERROR(cudaStreamEndCapture(enc.stream(), &graph));
-  std::unique_ptr<cudaGraph_t, void (*)(cudaGraph_t*)> graph_freer(
-      &graph, [](cudaGraph_t* p) { CHECK_CUDA_ERROR(cudaGraphDestroy(*p)); });
-  if (discard) {
-    return;
+  size_t num_nodes;
+  CHECK_CUDA_ERROR(cudaGraphGetNodes(graph, NULL, &num_nodes));
+  if (num_nodes == 1) {
+    cudaGraphNode_t captured_node;
+    CHECK_CUDA_ERROR(cudaGraphGetNodes(graph, &captured_node, &num_nodes));
+    CUDA_KERNEL_NODE_PARAMS params;
+    CHECK_CUDA_ERROR(cuGraphKernelNodeGetParams(captured_node, &params));
+    cudaGraphNode_t node;
+    CHECK_CUDA_ERROR(cuGraphAddKernelNode(&node, enc.graph_, NULL, 0, &params));
+    enc.insert_graph_dependencies(GraphNode{node, 'K'});
+  } else {
+    cudaGraphNode_t node;
+    CHECK_CUDA_ERROR(
+        cudaGraphAddChildGraphNode(&node, enc.graph_, NULL, 0, graph));
+    enc.insert_graph_dependencies(GraphNode{node, 'G'});
   }
-  enc.add_graph_node(graph);
+  CHECK_CUDA_ERROR(cudaGraphDestroy(graph));
 }
 
 CommandEncoder::ConcurrentContext::ConcurrentContext(CommandEncoder& enc)
@@ -184,11 +175,21 @@ void CommandEncoder::insert_graph_dependencies(std::vector<GraphNode> nodes) {
   }
 }
 
-CommandEncoder::CommandEncoder(Device& d)
-    : device_(d), stream_(d), graph_cache_(cuda_graph_cache_size()) {
+CommandEncoder::CommandEncoder(Device& d) : device_(d), stream_(d) {
   CHECK_CUDA_ERROR(cudaGraphCreate(&graph_, 0));
 }
 
+void clear_graphs(std::unordered_map<std::string, cudaGraphExec_t>& graphs) {
+  for (auto& [_, graph_exec] : graphs) {
+    CHECK_CUDA_ERROR(cudaGraphExecDestroy(graph_exec));
+  }
+  graphs.clear();
+}
+
+CommandEncoder::~CommandEncoder() {
+  clear_graphs(graph_cache_);
+}
+
 void CommandEncoder::add_completed_handler(std::function<void()> task) {
   worker_.add_task(std::move(task));
 }
@@ -222,7 +223,10 @@ void CommandEncoder::add_kernel_node(
   kernel_params.blockDim = block_dim;
   kernel_params.kernelParams = params;
   kernel_params.sharedMemBytes = smem_bytes;
-  add_kernel_node(kernel_params);
+  cudaGraphNode_t node;
+  CHECK_CUDA_ERROR(
+      cudaGraphAddKernelNode(&node, graph_, NULL, 0, &kernel_params));
+  insert_graph_dependencies(GraphNode{node, 'K'});
 }
 
 void CommandEncoder::add_kernel_node(
@@ -241,29 +245,13 @@ void CommandEncoder::add_kernel_node(
   kernel_params.blockDimZ = block_dim.z;
   kernel_params.kernelParams = params;
   kernel_params.sharedMemBytes = smem_bytes;
-  add_kernel_node(kernel_params);
-}
-
-void CommandEncoder::add_kernel_node(const cudaKernelNodeParams& params) {
-  cudaGraphNode_t node;
-  CHECK_CUDA_ERROR(cudaGraphAddKernelNode(&node, graph_, NULL, 0, &params));
-  insert_graph_dependencies(GraphNode{node, 'K'});
-}
-
-void CommandEncoder::add_kernel_node(const CUDA_KERNEL_NODE_PARAMS& params) {
   CUgraphNode node;
-  CHECK_CUDA_ERROR(cuGraphAddKernelNode(&node, graph_, NULL, 0, &params));
+  CHECK_CUDA_ERROR(
+      cuGraphAddKernelNode(&node, graph_, NULL, 0, &kernel_params));
   insert_graph_dependencies(GraphNode{node, 'K'});
 }
 
-void CommandEncoder::add_graph_node(cudaGraph_t child) {
-  cudaGraphNode_t node;
-  CHECK_CUDA_ERROR(cudaGraphAddChildGraphNode(&node, graph_, NULL, 0, child));
-  insert_graph_dependencies(GraphNode{node, 'G'});
-}
-
 void CommandEncoder::commit() {
-  nvtx3::scoped_range r("CommandEncoder::commit");
   if (!temporaries_.empty()) {
     add_completed_handler([temporaries = std::move(temporaries_)]() {});
   }
@@ -280,7 +268,7 @@ void CommandEncoder::commit() {
     graph_key_ += ".";
     graph_key_ += std::to_string(empty_node_count_);
 
-    CudaGraphExec& graph_exec = graph_cache_[graph_key_];
+    cudaGraphExec_t& graph_exec = graph_cache_[graph_key_];
 
     if (graph_exec != nullptr) {
       cudaGraphExecUpdateResult update_result;
@@ -294,19 +282,25 @@ void CommandEncoder::commit() {
 #endif // CUDART_VERSION >= 12000
       if (update_result != cudaGraphExecUpdateSuccess) {
         cudaGetLastError(); // reset error
-        graph_exec.reset();
+        CHECK_CUDA_ERROR(cudaGraphExecDestroy(graph_exec));
+        graph_exec = nullptr;
       }
     }
     if (graph_exec == nullptr) {
-      graph_exec.instantiate(graph_);
+      CHECK_CUDA_ERROR(
+          cudaGraphInstantiate(&graph_exec, graph_, NULL, NULL, 0));
     }
     device_.make_current();
     CHECK_CUDA_ERROR(cudaGraphLaunch(graph_exec, stream_));
 
+    // TODO smarter cache policy
+    if (graph_cache_.size() > cuda_graph_cache_size()) {
+      clear_graphs(graph_cache_);
+    }
+
     // Reset state
     node_count_ = 0;
     graph_node_count_ = 0;
-    empty_node_count_ = 0;
     from_nodes_.clear();
     to_nodes_.clear();
     graph_key_.clear();
@@ -316,6 +310,7 @@ void CommandEncoder::commit() {
   }
 
   // Put completion handlers in a batch.
+  worker_.end_batch();
   worker_.commit(stream_);
 }
 
@@ -324,6 +319,7 @@ void CommandEncoder::synchronize() {
   auto p = std::make_shared<std::promise<void>>();
   std::future<void> f = p->get_future();
   add_completed_handler([p = std::move(p)]() { p->set_value(); });
+  worker_.end_batch();
   commit();
   f.wait();
 }
@@ -341,4 +337,6 @@ CommandEncoder& get_command_encoder(Stream s) {
   return device(s.device).get_command_encoder(s);
 }
 
-} // namespace mlx::core::cu
+} // namespace cu
+
+} // namespace mlx::core

mlx/backend/cuda/device.h

@@ -3,13 +3,11 @@
 #pragma once
 
 #include "mlx/array.h"
-#include "mlx/backend/cuda/lru_cache.h"
 #include "mlx/backend/cuda/worker.h"
 #include "mlx/stream.h"
 
 #include <cublasLt.h>
 #include <cuda.h>
-#include <cudnn.h>
 #include <thrust/execution_policy.h>
 
 #include <unordered_map>
@@ -23,7 +21,6 @@ class CommandEncoder {
     ~CaptureContext();
     cudaGraph_t graph;
     CommandEncoder& enc;
-    bool discard{false};
   };
   struct ConcurrentContext {
     ConcurrentContext(CommandEncoder& enc);
@@ -32,6 +29,7 @@ class CommandEncoder {
   };
 
   explicit CommandEncoder(Device& d);
+  ~CommandEncoder();
 
   CommandEncoder(const CommandEncoder&) = delete;
   CommandEncoder& operator=(const CommandEncoder&) = delete;
@@ -76,11 +74,6 @@ class CommandEncoder {
       uint32_t smem_bytes,
       void** params);
 
-  // Low-level graph helpers.
-  void add_kernel_node(const cudaKernelNodeParams& params);
-  void add_kernel_node(const CUDA_KERNEL_NODE_PARAMS& params);
-  void add_graph_node(cudaGraph_t child);
-
   void add_temporary(const array& arr) {
     temporaries_.push_back(arr.data_shared_ptr());
   }
@@ -89,10 +82,6 @@ class CommandEncoder {
   void maybe_commit();
   void commit();
 
-  Device& device() {
-    return device_;
-  }
-
   CudaStream& stream() {
     return stream_;
   }
@@ -126,7 +115,7 @@ class CommandEncoder {
   std::string graph_key_;
   std::vector<GraphNode> concurrent_nodes_;
   std::vector<std::shared_ptr<array::Data>> temporaries_;
-  LRUCache<std::string, CudaGraphExec> graph_cache_;
+  std::unordered_map<std::string, cudaGraphExec_t> graph_cache_;
   std::vector<std::uintptr_t> active_deps_;
   std::vector<std::uintptr_t> active_outputs_;
   std::unordered_map<std::uintptr_t, GraphNode> node_map_;
@@ -157,16 +146,12 @@ class Device {
   cublasLtHandle_t lt_handle() const {
     return lt_;
   }
-  cudnnHandle_t cudnn_handle() const {
-    return cudnn_;
-  }
 
  private:
   int device_;
   int compute_capability_major_;
   int compute_capability_minor_;
   cublasLtHandle_t lt_;
-  cudnnHandle_t cudnn_;
   std::unordered_map<int, CommandEncoder> encoders_;
 };
 

mlx/backend/cuda/device/arange.cuh

@@ -0,0 +1,15 @@
+// Copyright © 2025 Apple Inc.
+
+namespace mlx::core::cu {
+
+template <typename T>
+struct Arange {
+  const T start;
+  const T step;
+
+  __device__ T operator()(uint32_t i) const {
+    return start + i * step;
+  }
+};
+
+} // namespace mlx::core::cu

mlx/backend/cuda/device/atomic_ops.cuh

@@ -49,7 +49,11 @@ inline __device__ void atomic_add(__half* out, __half val) {
 }
 
 inline __device__ void atomic_add(complex64_t* out, complex64_t val) {
+#if __CUDA_ARCH__ < 900
   atomic_add_general(out, val);
+#else
+  atomicAdd(out, val);
+#endif
 }
 
 inline __device__ void atomic_add(__nv_bfloat16* out, __nv_bfloat16 val) {

mlx/backend/cuda/device/utils.cuh

@@ -32,118 +32,21 @@ using Strides = cuda::std::array<int64_t, MAX_NDIM>;
 template <typename T, int N>
 struct alignas(sizeof(T) * N) AlignedVector {
   T val[N];
-
-  __device__ T& operator[](int i) {
-    return val[i];
-  }
-
-  __device__ T operator[](int i) const {
-    return val[i];
-  }
 };
 
 template <int N, typename T>
-inline __host__ __device__ bool is_aligned(T* x) {
-  return (reinterpret_cast<uintptr_t>(x) % (N * sizeof(T))) == 0;
-}
-
-template <int N, typename T>
-inline __device__ AlignedVector<T, N> unsafe_load_vector(
+inline __device__ AlignedVector<T, N> load_vector(
     const T* ptr,
     uint32_t offset) {
   auto* from = reinterpret_cast<const AlignedVector<T, N>*>(ptr);
   return from[offset];
 }
 
-template <int N, typename T>
-inline __device__ AlignedVector<T, N> load_vector(
-    const T* ptr,
-    uint32_t offset) {
-  if (is_aligned<N>(ptr)) {
-    auto* from = reinterpret_cast<const AlignedVector<T, N>*>(ptr);
-    return from[offset];
-  } else {
-    AlignedVector<T, N> v;
-#pragma unroll
-    for (int i = 0; i < N; ++i) {
-      v[i] = ptr[offset * N + i];
-    }
-    return v;
-  }
-}
-
-template <int N, typename T, typename SizeT>
-inline __device__ AlignedVector<T, N>
-load_vector(const T* ptr, uint32_t offset, SizeT size, T fallback) {
-  if (is_aligned<N>(ptr) && (offset + 1) * N <= size) {
-    auto* from = reinterpret_cast<const AlignedVector<T, N>*>(ptr);
-    return from[offset];
-  } else {
-    AlignedVector<T, N> v;
-#pragma unroll
-    for (int i = 0; i < N; ++i) {
-      v[i] = (N * offset + i) < size ? ptr[offset * N + i] : fallback;
-    }
-    return v;
-  }
-}
-
-template <int N, typename T, typename SizeT>
-inline __device__ AlignedVector<T, N> load_vector(
-    const T* ptr,
-    uint32_t offset,
-    SizeT size,
-    int64_t stride,
-    T fallback) {
-  if (is_aligned<N>(ptr) && stride == 1 && (offset + 1) * N <= size) {
-    auto* from = reinterpret_cast<const AlignedVector<T, N>*>(ptr);
-    return from[offset];
-  } else {
-    AlignedVector<T, N> v;
-#pragma unroll
-    for (int i = 0; i < N; ++i) {
-      v[i] =
-          (N * offset + i) < size ? ptr[stride * (offset * N + i)] : fallback;
-    }
-    return v;
-  }
-}
-
-template <int N, typename T>
-inline __device__ void
-unsafe_store_vector(T* ptr, uint32_t offset, const AlignedVector<T, N>& vec) {
-  auto* to = reinterpret_cast<AlignedVector<T, N>*>(ptr);
-  to[offset] = vec;
-}
-
 template <int N, typename T>
 inline __device__ void
 store_vector(T* ptr, uint32_t offset, const AlignedVector<T, N>& vec) {
-  if (is_aligned<N>(ptr)) {
-    auto* to = reinterpret_cast<AlignedVector<T, N>*>(ptr);
-    to[offset] = vec;
-  } else {
-#pragma unroll
-    for (int i = 0; i < N; ++i) {
-      ptr[offset * N + i] = vec[i];
-    }
-  }
-}
-
-template <int N, typename T, typename SizeT>
-inline __device__ void store_vector(
-    T* ptr,
-    uint32_t offset,
-    const AlignedVector<T, N>& vec,
-    SizeT size) {
-  if (is_aligned<N>(ptr) && (offset + 1) * N <= size) {
-    auto* to = reinterpret_cast<AlignedVector<T, N>*>(ptr);
-    to[offset] = vec;
-  } else {
-    for (int i = 0; (offset * N + i) < size && i < N; ++i) {
-      ptr[offset * N + i] = vec[i];
-    }
-  }
+  auto* to = reinterpret_cast<AlignedVector<T, N>*>(ptr);
+  to[offset] = vec;
 }
 
 ///////////////////////////////////////////////////////////////////////////////
@@ -301,8 +204,20 @@ inline __host__ __device__ cuda::std::tuple<IdxT, IdxT, IdxT> elem_to_loc_nd(
   return cuda::std::make_tuple(a_loc, b_loc, c_loc);
 }
 
+// Optimized version when ndim is larger than 4.
 template <typename IdxT = int64_t>
-inline __host__ __device__ cuda::std::tuple<IdxT, IdxT> elem_to_loc(
+inline __host__ __device__ IdxT
+elem_to_loc_4d(IdxT elem, const int* shape, const int64_t* strides, int ndim) {
+  IdxT loc = 0;
+  for (int i = ndim - 1; i >= 0; --i) {
+    loc += (elem % shape[i]) * IdxT(strides[i]);
+    elem /= shape[i];
+  }
+  return loc;
+}
+
+template <typename IdxT = int64_t>
+inline __host__ __device__ cuda::std::tuple<IdxT, IdxT> elem_to_loc_4d(
     IdxT elem,
     const int* shape,
     const int64_t* a_strides,
@@ -320,7 +235,7 @@ inline __host__ __device__ cuda::std::tuple<IdxT, IdxT> elem_to_loc(
 }
 
 template <typename IdxT = int64_t>
-inline __host__ __device__ cuda::std::tuple<IdxT, IdxT, IdxT> elem_to_loc(
+inline __host__ __device__ cuda::std::tuple<IdxT, IdxT, IdxT> elem_to_loc_4d(
     IdxT elem,
     const int* shape,
     const int64_t* a_strides,

mlx/backend/cuda/distributed.cu

@@ -1,51 +0,0 @@
-// Copyright © 2025 Apple Inc.
-
-#include "mlx/backend/cuda/device.h"
-#include "mlx/backend/cuda/kernel_utils.cuh"
-#include "mlx/distributed/primitives.h"
-#include "mlx/primitives.h"
-
-#include <cassert>
-
-namespace mlx::core {
-namespace distributed {
-void AllReduce::eval_gpu(
-    const std::vector<array>& inputs,
-    std::vector<array>& outputs) {
-  assert(inputs.size() == 1);
-  assert(outputs.size() == 1);
-
-  auto& input = inputs[0];
-  auto& output = outputs[0];
-
-  auto& encoder = cu::get_command_encoder(stream());
-
-  if (input.is_donatable()) {
-    output.copy_shared_buffer(input);
-  } else {
-    output.set_data(allocator::malloc(output.nbytes()));
-  }
-
-  encoder.set_input_array(input);
-  encoder.set_output_array(output);
-
-  auto capture = encoder.capture_context();
-  auto& s = stream();
-
-  switch (reduce_type_) {
-    case Sum:
-      distributed::detail::all_sum(group(), input, output, s);
-      break;
-    case Max:
-      distributed::detail::all_max(group(), input, output, s);
-      break;
-    case Min:
-      distributed::detail::all_min(group(), input, output, s);
-      break;
-    default:
-      throw std::runtime_error(
-          "Only all reduce sum, max, and min are supported.");
-  }
-}
-} // namespace distributed
-} // namespace mlx::core

mlx/backend/cuda/eval.cpp

@@ -19,6 +19,8 @@ void new_stream(Stream s) {
   cudaFree(nullptr);
   // Ensure the static stream objects get created.
   cu::get_command_encoder(s);
+  // The main thread is safe to free buffers.
+  cu::allocator().register_this_thread();
 }
 
 void eval(array& arr) {
@@ -36,15 +38,18 @@ void eval(array& arr) {
 
   auto& encoder = cu::get_command_encoder(arr.primitive().stream());
   // Keep used buffers alive until kernel finishes running.
+  std::unordered_set<std::shared_ptr<array::Data>> buffers;
   for (auto& in : arr.inputs()) {
-    // Except for the donated one.
-    if (in.data_shared_ptr() != arr.data_shared_ptr()) {
-      encoder.add_temporary(in);
-    }
+    buffers.insert(in.data_shared_ptr());
   }
   for (auto& s : arr.siblings()) {
-    encoder.add_temporary(s);
+    buffers.insert(s.data_shared_ptr());
   }
+  // Remove the output if it was donated to by an input.
+  if (auto it = buffers.find(arr.data_shared_ptr()); it != buffers.end()) {
+    buffers.erase(it);
+  }
+  encoder.add_completed_handler([buffers = std::move(buffers)]() {});
   encoder.maybe_commit();
 }
 

mlx/backend/cuda/event.cu

@@ -110,26 +110,24 @@ __global__ void event_signal_kernel(SharedEvent::Atomic* ac, uint64_t value) {
   event_signal(ac, value);
 }
 
-SharedEvent::Atomic* to_atomic(std::shared_ptr<Buffer> buf) {
-  return static_cast<SharedEvent::Atomic*>(buf->raw_ptr());
-}
-
 SharedEvent::SharedEvent() {
-  buf_ = std::shared_ptr<Buffer>(
-      new Buffer{allocator().malloc(sizeof(Atomic))}, [](Buffer* ptr) {
-        allocator().free(*ptr);
-        delete ptr;
-      });
-  *static_cast<uint64_t*>(buf_->raw_ptr()) = 0;
+  // Allocate cuda::atomic on managed memory.
+  Atomic* ac;
+  CHECK_CUDA_ERROR(cudaMallocManaged(&ac, sizeof(Atomic)));
+  new (ac) Atomic(0);
+  ac_ = std::shared_ptr<Atomic>(ac, [](Atomic* ptr) {
+    ptr->~Atomic();
+    allocator().cuda_free(ptr);
+  });
 }
 
 void SharedEvent::wait(uint64_t value) {
   nvtx3::scoped_range r("cu::SharedEvent::wait");
-  event_wait(to_atomic(buf_), value);
+  event_wait(ac_.get(), value);
 }
 
 void SharedEvent::wait(cudaStream_t stream, uint64_t value) {
-  event_wait_kernel<<<1, 1, 0, stream>>>(to_atomic(buf_), value);
+  event_wait_kernel<<<1, 1, 0, stream>>>(ac_.get(), value);
 }
 
 void SharedEvent::wait(Stream s, uint64_t value) {
@@ -140,17 +138,17 @@ void SharedEvent::wait(Stream s, uint64_t value) {
     auto& encoder = get_command_encoder(s);
     encoder.commit();
     wait(encoder.stream(), value);
-    encoder.add_completed_handler([buf = buf_]() {});
+    encoder.add_completed_handler([ac = ac_]() {});
   }
 }
 
 void SharedEvent::signal(uint64_t value) {
   nvtx3::scoped_range r("cu::SharedEvent::signal");
-  event_signal(to_atomic(buf_), value);
+  event_signal(ac_.get(), value);
 }
 
 void SharedEvent::signal(cudaStream_t stream, uint64_t value) {
-  event_signal_kernel<<<1, 1, 0, stream>>>(to_atomic(buf_), value);
+  event_signal_kernel<<<1, 1, 0, stream>>>(ac_.get(), value);
 }
 
 void SharedEvent::signal(Stream s, uint64_t value) {
@@ -164,18 +162,18 @@ void SharedEvent::signal(Stream s, uint64_t value) {
     auto& encoder = get_command_encoder(s);
     encoder.commit();
     signal(encoder.stream(), value);
-    encoder.add_completed_handler([buf = buf_]() {});
+    encoder.add_completed_handler([ac = ac_]() {});
   }
 }
 
 bool SharedEvent::is_signaled(uint64_t value) const {
   nvtx3::scoped_range r("cu::SharedEvent::is_signaled");
-  return to_atomic(buf_)->load() >= value;
+  return ac_->load() >= value;
 }
 
 uint64_t SharedEvent::value() const {
   nvtx3::scoped_range r("cu::SharedEvent::value");
-  return to_atomic(buf_)->load();
+  return ac_->load();
 }
 
 } // namespace cu

mlx/backend/cuda/event.h

@@ -2,7 +2,6 @@
 
 #pragma once
 
-#include "mlx/allocator.h"
 #include "mlx/stream.h"
 
 #include <cuda_runtime.h>
@@ -56,8 +55,12 @@ class SharedEvent {
   bool is_signaled(uint64_t value) const;
   uint64_t value() const;
 
+  const std::shared_ptr<Atomic>& atomic() const {
+    return ac_;
+  }
+
  private:
-  std::shared_ptr<mlx::core::allocator::Buffer> buf_;
+  std::shared_ptr<Atomic> ac_;
 };
 
 } // namespace mlx::core::cu

mlx/backend/cuda/gemms/cublas_batched_gemm_12_0.cpp

@@ -1,73 +0,0 @@
-// Copyright © 2025 Apple Inc.
-
-#include "mlx/backend/common/utils.h"
-#include "mlx/backend/cuda/device.h"
-#include "mlx/backend/cuda/gemms/cublas_gemm.h"
-
-namespace mlx::core::cu {
-
-void Matmul::run_batched(
-    cu::CommandEncoder& encoder,
-    array& out,
-    const array& a,
-    const array& b,
-    const mlx::core::Shape& batch_shape,
-    const mlx::core::Strides& a_batch_strides,
-    const mlx::core::Strides& b_batch_strides) {
-  encoder.set_input_array(a);
-  encoder.set_input_array(b);
-  encoder.set_output_array(out);
-  auto nbatch = out.size() / (M_ * N_ * batch_shape.back());
-  ContiguousIterator a_it(batch_shape, a_batch_strides, batch_shape.size() - 1);
-  ContiguousIterator b_it(batch_shape, b_batch_strides, batch_shape.size() - 1);
-  auto concurrent = encoder.concurrent_context();
-  for (size_t i = 0; i < nbatch; ++i) {
-    run_impl(
-        encoder,
-        out.data<int8_t>() + out.itemsize() * i * batch_shape.back() * M_ * N_,
-        a.data<int8_t>() + a.itemsize() * a_it.loc,
-        b.data<int8_t>() + b.itemsize() * b_it.loc,
-        nullptr);
-    a_it.step();
-    b_it.step();
-  }
-}
-
-void Matmul::run_batched(
-    cu::CommandEncoder& encoder,
-    array& out,
-    const array& a,
-    const array& b,
-    const array& c,
-    const mlx::core::Shape& batch_shape,
-    const mlx::core::Strides& a_batch_strides,
-    const mlx::core::Strides& b_batch_strides,
-    const mlx::core::Strides& c_batch_strides,
-    float alpha,
-    float beta) {
-  encoder.set_input_array(a);
-  encoder.set_input_array(b);
-  encoder.set_input_array(c);
-  encoder.set_output_array(out);
-
-  auto nbatch = out.size() / (M_ * N_ * batch_shape.back());
-  ContiguousIterator a_it(batch_shape, a_batch_strides, batch_shape.size() - 1);
-  ContiguousIterator b_it(batch_shape, b_batch_strides, batch_shape.size() - 1);
-  ContiguousIterator c_it(batch_shape, c_batch_strides, batch_shape.size() - 1);
-  auto concurrent = encoder.concurrent_context();
-  for (size_t i = 0; i < nbatch; ++i) {
-    run_impl(
-        encoder,
-        out.data<int8_t>() + out.itemsize() * i * batch_shape.back() * M_ * N_,
-        a.data<int8_t>() + a.itemsize() * a_it.loc,
-        b.data<int8_t>() + b.itemsize() * b_it.loc,
-        c.data<int8_t>() + c.itemsize() * c_it.loc,
-        alpha,
-        beta);
-    a_it.step();
-    b_it.step();
-    c_it.step();
-  }
-}
-
-} // namespace mlx::core::cu

mlx/backend/cuda/gemms/cublas_batched_gemm_12_9.cu

@@ -1,208 +0,0 @@
-// Copyright © 2025 Apple Inc.
-
-#include "mlx/backend/cuda/device.h"
-#include "mlx/backend/cuda/gemms/cublas_gemm.h"
-#include "mlx/backend/cuda/kernel_utils.cuh"
-
-#include <cooperative_groups.h>
-
-namespace mlx::core::cu {
-
-namespace cg = cooperative_groups;
-
-__global__ void set_mm_device_pointers(
-    int8_t** pointers,
-    int8_t* a_start,
-    int8_t* b_start,
-    int8_t* out_start,
-    int item_size,
-    const __grid_constant__ Shape batch_shape,
-    const __grid_constant__ Strides a_batch_strides,
-    const __grid_constant__ Strides b_batch_strides,
-    int64_t batch_stride,
-    int batch_ndim,
-    int batch_count) {
-  auto index = cg::this_grid().thread_rank();
-  if (index >= batch_count) {
-    return;
-  }
-  auto [a_offset, b_offset] = elem_to_loc(
-      index,
-      batch_shape.data(),
-      a_batch_strides.data(),
-      b_batch_strides.data(),
-      batch_ndim);
-  pointers[index] = a_start + item_size * a_offset;
-  pointers[index + batch_count] = b_start + item_size * b_offset;
-  pointers[index + 2 * batch_count] =
-      out_start + item_size * index * batch_stride;
-}
-
-__global__ void set_addmm_device_pointers(
-    int8_t** pointers,
-    int8_t* a_start,
-    int8_t* b_start,
-    int8_t* c_start,
-    int8_t* out_start,
-    int item_size,
-    const __grid_constant__ Shape batch_shape,
-    const __grid_constant__ Strides a_batch_strides,
-    const __grid_constant__ Strides b_batch_strides,
-    const __grid_constant__ Strides c_batch_strides,
-    int64_t batch_stride,
-    int batch_ndim,
-    int batch_count) {
-  auto index = cg::this_grid().thread_rank();
-  if (index >= batch_count) {
-    return;
-  }
-  auto [a_offset, b_offset, c_offset] = elem_to_loc(
-      index,
-      batch_shape.data(),
-      a_batch_strides.data(),
-      b_batch_strides.data(),
-      c_batch_strides.data(),
-      batch_ndim);
-  pointers[index] = a_start + item_size * a_offset;
-  pointers[index + batch_count] = b_start + item_size * b_offset;
-  pointers[index + 2 * batch_count] = c_start + item_size * c_offset;
-  pointers[index + 3 * batch_count] =
-      out_start + item_size * index * batch_stride;
-}
-
-void set_pointer_mode(cublasLtMatrixLayout_t desc, int batch_count) {
-  auto batch_mode = CUBLASLT_BATCH_MODE_POINTER_ARRAY;
-  CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutSetAttribute(
-      desc,
-      CUBLASLT_MATRIX_LAYOUT_BATCH_MODE,
-      &batch_mode,
-      sizeof(batch_mode)));
-  CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutSetAttribute(
-      desc, CUBLASLT_MATRIX_LAYOUT_BATCH_COUNT, &batch_count, sizeof(int32_t)));
-}
-
-void Matmul::run_batched(
-    cu::CommandEncoder& encoder,
-    array& out,
-    const array& a,
-    const array& b,
-    const mlx::core::Shape& batch_shape,
-    const mlx::core::Strides& a_batch_strides,
-    const mlx::core::Strides& b_batch_strides) {
-  auto batch_count = out.size() / (M_ * N_);
-  set_pointer_mode(a_desc_, batch_count);
-  set_pointer_mode(b_desc_, batch_count);
-  set_pointer_mode(out_desc_, batch_count);
-
-  // Launch kernel to set device offsets
-  auto pointers = array(
-      allocator::malloc(batch_count * sizeof(uint64_t) * 3),
-      {static_cast<int>(batch_count * 3)},
-      uint64);
-
-  encoder.add_temporary(pointers);
-  int block_size = 512;
-  encoder.set_output_array(pointers);
-
-  encoder.add_kernel_node(
-      cu::set_mm_device_pointers,
-      cuda::ceil_div(pointers.size(), block_size),
-      block_size,
-      0,
-      pointers.data<int8_t*>(),
-      a.data<int8_t>(),
-      b.data<int8_t>(),
-      out.data<int8_t>(),
-      static_cast<int>(out.dtype().size()),
-      const_param(batch_shape),
-      const_param(a_batch_strides),
-      const_param(b_batch_strides),
-      static_cast<int64_t>(M_) * N_,
-      static_cast<int>(batch_shape.size()),
-      batch_count);
-
-  // Run matmul
-  encoder.set_input_array(pointers);
-  encoder.set_input_array(a);
-  encoder.set_input_array(b);
-  encoder.set_output_array(out);
-
-  auto a_pointers = pointers.data<int8_t*>();
-  auto b_pointers = a_pointers + batch_count;
-  auto out_pointers = b_pointers + batch_count;
-  run_impl(
-      encoder,
-      reinterpret_cast<void*>(out_pointers),
-      reinterpret_cast<void*>(a_pointers),
-      reinterpret_cast<void*>(b_pointers),
-      nullptr);
-}
-
-void Matmul::run_batched(
-    cu::CommandEncoder& encoder,
-    array& out,
-    const array& a,
-    const array& b,
-    const array& c,
-    const mlx::core::Shape& batch_shape,
-    const mlx::core::Strides& a_batch_strides,
-    const mlx::core::Strides& b_batch_strides,
-    const mlx::core::Strides& c_batch_strides,
-    float alpha,
-    float beta) {
-  auto batch_count = out.size() / (M_ * N_);
-  set_pointer_mode(a_desc_, batch_count);
-  set_pointer_mode(b_desc_, batch_count);
-  set_pointer_mode(c_desc_, batch_count);
-  set_pointer_mode(out_desc_, batch_count);
-
-  // Launch kernel to set device offsets
-  auto pointers = array(
-      allocator::malloc(batch_count * sizeof(uint64_t) * 4),
-      {static_cast<int>(batch_count * 4)},
-      uint64);
-
-  encoder.add_temporary(pointers);
-  int block_size = 512;
-  encoder.set_output_array(pointers);
-  encoder.add_kernel_node(
-      cu::set_addmm_device_pointers,
-      cuda::ceil_div(pointers.size(), block_size),
-      block_size,
-      0,
-      pointers.data<int8_t*>(),
-      a.data<int8_t>(),
-      b.data<int8_t>(),
-      c.data<int8_t>(),
-      out.data<int8_t>(),
-      static_cast<int>(out.dtype().size()),
-      const_param(batch_shape),
-      const_param(a_batch_strides),
-      const_param(b_batch_strides),
-      const_param(c_batch_strides),
-      static_cast<int64_t>(M_) * N_,
-      static_cast<int>(batch_shape.size()),
-      batch_count);
-
-  // Run matmul
-  encoder.set_input_array(pointers);
-  encoder.set_input_array(a);
-  encoder.set_input_array(b);
-  encoder.set_input_array(c);
-  encoder.set_output_array(out);
-
-  auto a_pointers = pointers.data<int8_t*>();
-  auto b_pointers = a_pointers + batch_count;
-  auto c_pointers = b_pointers + batch_count;
-  auto out_pointers = c_pointers + batch_count;
-  run_impl(
-      encoder,
-      reinterpret_cast<void*>(out_pointers),
-      reinterpret_cast<void*>(a_pointers),
-      reinterpret_cast<void*>(b_pointers),
-      reinterpret_cast<void*>(c_pointers),
-      alpha,
-      beta);
-}
-
-} // namespace mlx::core::cu

mlx/backend/cuda/gemms/cublas_gemm.cpp

@@ -1,284 +0,0 @@
-// Copyright © 2025 Apple Inc.
-
-#include "mlx/backend/cuda/gemms/cublas_gemm.h"
-#include "mlx/backend/cuda/device.h"
-#include "mlx/dtype_utils.h"
-#include "mlx/utils.h"
-
-#include <fmt/format.h>
-
-namespace mlx::core::cu {
-
-struct CublasPreference {
-  CublasPreference(Device& device) {
-    // The recommended cublas workspace size is 4 MiB for pre-Hopper and 32 MiB
-    // for Hopper+:
-    // https://docs.nvidia.com/cuda/cublas/#cublassetworkspace
-    uint64_t MiB = 1024 * 1024;
-    uint64_t workspace_size =
-        device.compute_capability_major() >= 9 ? 32 * MiB : 4 * MiB;
-
-    CHECK_CUBLAS_ERROR(cublasLtMatmulPreferenceCreate(&pref_));
-    CHECK_CUBLAS_ERROR(cublasLtMatmulPreferenceSetAttribute(
-        pref_,
-        CUBLASLT_MATMUL_PREF_MAX_WORKSPACE_BYTES,
-        &workspace_size,
-        sizeof(uint64_t)));
-  }
-
-  ~CublasPreference() {
-    CHECK_CUBLAS_ERROR(cublasLtMatmulPreferenceDestroy(pref_));
-  }
-
-  cublasLtMatmulPreference_t pref_{nullptr};
-};
-
-cublasLtMatmulPreference_t cublas_preference(Device& device) {
-  static CublasPreference pref(device);
-  return pref.pref_;
-}
-
-cublasComputeType_t dtype_to_compute_type(Dtype dtype) {
-  switch (dtype) {
-    case float16:
-      return CUBLAS_COMPUTE_32F;
-    case bfloat16:
-      return CUBLAS_COMPUTE_32F;
-    case float32:
-      return mlx::core::env::enable_tf32() ? CUBLAS_COMPUTE_32F_FAST_TF32
-                                           : CUBLAS_COMPUTE_32F;
-    case float64:
-    case complex64:
-      return CUBLAS_COMPUTE_64F;
-    default:
-      throw std::runtime_error(fmt::format(
-          "Unsupported dtype in Matmul: {}.", dtype_to_string(dtype)));
-  }
-}
-
-cudaDataType_t dtype_to_cublas_type(Dtype dtype) {
-  switch (dtype) {
-    case float16:
-      return CUDA_R_16F;
-    case bfloat16:
-      return CUDA_R_16BF;
-    case float32:
-      return CUDA_R_32F;
-    case float64:
-      return CUDA_R_64F;
-    case complex64:
-      return CUDA_C_32F;
-    default:
-      throw std::runtime_error(fmt::format(
-          "Unsupported dtype in Matmul: {}.", dtype_to_string(dtype)));
-  }
-}
-
-cublasLtMatrixLayout_t create_matrix_layout(
-    cudaDataType_t type,
-    uint64_t rows,
-    uint64_t cols,
-    bool transposed,
-    int64_t ld,
-    int32_t batch_count,
-    int64_t batch_stride) {
-  cublasLtMatrixLayout_t desc;
-  CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutCreate(&desc, type, rows, cols, ld));
-  cublasLtOrder_t order = transposed ? CUBLASLT_ORDER_COL : CUBLASLT_ORDER_ROW;
-  CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutSetAttribute(
-      desc, CUBLASLT_MATRIX_LAYOUT_ORDER, &order, sizeof(cublasLtOrder_t)));
-  if (batch_count > 1) {
-    CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutSetAttribute(
-        desc,
-        CUBLASLT_MATRIX_LAYOUT_BATCH_COUNT,
-        &batch_count,
-        sizeof(int32_t)));
-    CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutSetAttribute(
-        desc,
-        CUBLASLT_MATRIX_LAYOUT_STRIDED_BATCH_OFFSET,
-        &batch_stride,
-        sizeof(int64_t)));
-  }
-  return desc;
-}
-
-Matmul::Matmul(
-    Device& device,
-    Dtype dtype,
-    bool a_transposed,
-    uint64_t a_rows,
-    uint64_t a_cols,
-    int64_t lda,
-    bool b_transposed,
-    uint64_t b_rows,
-    uint64_t b_cols,
-    int64_t ldb,
-    int32_t batch_count,
-    int64_t a_batch_stride,
-    int64_t b_batch_stride)
-    : handle_(device.lt_handle()),
-      pref_(cublas_preference(device)),
-      M_(a_rows),
-      N_(b_cols) {
-  heuristic_.state = CUBLAS_STATUS_NOT_INITIALIZED;
-
-  auto scale_type = dtype_to_cublas_type(dtype);
-  if (dtype == bfloat16 || dtype == float16) {
-    scale_type = CUDA_R_32F;
-  }
-  CHECK_CUBLAS_ERROR(cublasLtMatmulDescCreate(
-      &matmul_desc_, dtype_to_compute_type(dtype), scale_type));
-  int32_t pointer_mode = CUBLASLT_POINTER_MODE_HOST;
-  CHECK_CUBLAS_ERROR(cublasLtMatmulDescSetAttribute(
-      matmul_desc_,
-      CUBLASLT_MATMUL_DESC_POINTER_MODE,
-      &pointer_mode,
-      sizeof(int32_t)));
-  cublasOperation_t op = CUBLAS_OP_N;
-  CHECK_CUBLAS_ERROR(cublasLtMatmulDescSetAttribute(
-      matmul_desc_,
-      CUBLASLT_MATMUL_DESC_TRANSA,
-      &op,
-      sizeof(cublasOperation_t)));
-  CHECK_CUBLAS_ERROR(cublasLtMatmulDescSetAttribute(
-      matmul_desc_,
-      CUBLASLT_MATMUL_DESC_TRANSB,
-      &op,
-      sizeof(cublasOperation_t)));
-
-  auto type = dtype_to_cublas_type(dtype);
-  a_desc_ = create_matrix_layout(
-      type, a_rows, a_cols, a_transposed, lda, batch_count, a_batch_stride);
-  b_desc_ = create_matrix_layout(
-      type, b_rows, b_cols, b_transposed, ldb, batch_count, b_batch_stride);
-  out_desc_ = create_matrix_layout(
-      type, a_rows, b_cols, false, b_cols, batch_count, a_rows * b_cols);
-}
-
-Matmul::Matmul(
-    Device& device,
-    Dtype dtype,
-    bool a_transposed,
-    uint64_t a_rows,
-    uint64_t a_cols,
-    int64_t lda,
-    bool b_transposed,
-    uint64_t b_rows,
-    uint64_t b_cols,
-    int64_t ldb,
-    int64_t ldc,
-    int32_t batch_count,
-    int64_t a_batch_stride,
-    int64_t b_batch_stride,
-    int64_t c_batch_stride)
-    : Matmul(
-          device,
-          dtype,
-          a_transposed,
-          a_rows,
-          a_cols,
-          lda,
-          b_transposed,
-          b_rows,
-          b_cols,
-          ldb,
-          batch_count,
-          a_batch_stride,
-          b_batch_stride) {
-  auto type = dtype_to_cublas_type(dtype);
-  c_desc_ = create_matrix_layout(
-      type, a_rows, b_cols, false, ldc, batch_count, c_batch_stride);
-}
-
-Matmul::~Matmul() {
-  CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutDestroy(a_desc_));
-  CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutDestroy(b_desc_));
-  CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutDestroy(c_desc_));
-  CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutDestroy(out_desc_));
-  CHECK_CUBLAS_ERROR(cublasLtMatmulDescDestroy(matmul_desc_));
-}
-
-void Matmul::run_impl(
-    cu::CommandEncoder& encoder,
-    void* out,
-    const void* a,
-    const void* b,
-    const void* c,
-    float alpha /* = 1 */,
-    float beta /* = 0 */) {
-  if (heuristic_.state != CUBLAS_STATUS_SUCCESS) {
-    int ret = 0;
-    CHECK_CUBLAS_ERROR(cublasLtMatmulAlgoGetHeuristic(
-        handle_,
-        matmul_desc_,
-        a_desc_,
-        b_desc_,
-        c ? c_desc_ : out_desc_,
-        out_desc_,
-        pref_,
-        1,
-        &heuristic_,
-        &ret));
-    if (ret == 0) {
-      throw std::runtime_error("Can not find algorithm for matmul.");
-    }
-  }
-
-  void* workspace_ptr = nullptr;
-  if (heuristic_.workspaceSize > 0) {
-    // Ensure workspace is 256-byte aligned
-    int nbytes = cuda::ceil_div(heuristic_.workspaceSize, 256) * 256;
-    array workspace(
-        allocator::malloc(nbytes),
-        {static_cast<int>(heuristic_.workspaceSize)},
-        int8);
-    encoder.add_temporary(workspace);
-    workspace_ptr = workspace.data<void>();
-  }
-
-  auto capture = encoder.capture_context();
-  CHECK_CUBLAS_ERROR(cublasLtMatmul(
-      handle_,
-      matmul_desc_,
-      &alpha,
-      a,
-      a_desc_,
-      b,
-      b_desc_,
-      &beta,
-      c ? c : out,
-      c ? c_desc_ : out_desc_,
-      out,
-      out_desc_,
-      &heuristic_.algo,
-      workspace_ptr,
-      heuristic_.workspaceSize,
-      encoder.stream()));
-}
-
-void Matmul::run(
-    cu::CommandEncoder& encoder,
-    array& out,
-    const array& a,
-    const array& b,
-    const std::optional<array>& c /* = std::nullopt */,
-    float alpha /* = 1 */,
-    float beta /* = 0 */) {
-  encoder.set_input_array(a);
-  encoder.set_input_array(b);
-  if (c) {
-    encoder.set_input_array(*c);
-  }
-  encoder.set_output_array(out);
-
-  run_impl(
-      encoder,
-      out.data<void>(),
-      a.data<void>(),
-      b.data<void>(),
-      c ? c->data<void>() : nullptr,
-      alpha,
-      beta);
-}
-
-} // namespace mlx::core::cu

mlx/backend/cuda/gemms/cublas_gemm.h

@@ -1,100 +0,0 @@
-// Copyright © 2025 Apple Inc.
-#pragma once
-
-#include "mlx/array.h"
-#include "mlx/backend/cuda/device.h"
-
-#include <cublasLt.h>
-#include <optional>
-
-namespace mlx::core::cu {
-class Matmul {
- public:
-  Matmul(
-      Device& device,
-      Dtype dtype,
-      bool a_transposed,
-      uint64_t a_rows,
-      uint64_t a_cols,
-      int64_t lda,
-      bool b_transposed,
-      uint64_t b_rows,
-      uint64_t b_cols,
-      int64_t ldb,
-      int32_t batch_count,
-      int64_t a_batch_stride,
-      int64_t b_batch_stride);
-
-  Matmul(
-      Device& device,
-      Dtype dtype,
-      bool a_transposed,
-      uint64_t a_rows,
-      uint64_t a_cols,
-      int64_t lda,
-      bool b_transposed,
-      uint64_t b_rows,
-      uint64_t b_cols,
-      int64_t ldb,
-      int64_t ldc,
-      int32_t batch_count,
-      int64_t a_batch_stride,
-      int64_t b_batch_stride,
-      int64_t c_batch_stride);
-
-  ~Matmul();
-
-  void run(
-      cu::CommandEncoder& encoder,
-      array& out,
-      const array& a,
-      const array& b,
-      const std::optional<array>& c = std::nullopt,
-      float alpha = 1,
-      float beta = 0);
-
-  void run_batched(
-      cu::CommandEncoder& encoder,
-      array& out,
-      const array& a,
-      const array& b,
-      const mlx::core::Shape& batch_shape,
-      const mlx::core::Strides& a_batch_strides,
-      const mlx::core::Strides& b_batch_strides);
-
-  void run_batched(
-      cu::CommandEncoder& encoder,
-      array& out,
-      const array& a,
-      const array& b,
-      const array& c,
-      const mlx::core::Shape& batch_shape,
-      const mlx::core::Strides& a_batch_strides,
-      const mlx::core::Strides& b_batch_strides,
-      const mlx::core::Strides& c_batch_strides,
-      float alpha,
-      float beta);
-
- private:
-  void run_impl(
-      cu::CommandEncoder& encoder,
-      void* out,
-      const void* a,
-      const void* b,
-      const void* c,
-      float alpha = 1,
-      float beta = 0);
-
-  uint64_t M_;
-  uint64_t N_;
-  cublasLtMatmulPreference_t pref_{nullptr};
-  cublasLtHandle_t handle_{nullptr};
-  cublasLtMatmulDesc_t matmul_desc_{nullptr};
-  cublasLtMatrixLayout_t a_desc_{nullptr};
-  cublasLtMatrixLayout_t b_desc_{nullptr};
-  cublasLtMatrixLayout_t c_desc_{nullptr};
-  cublasLtMatrixLayout_t out_desc_{nullptr};
-  cublasLtMatmulHeuristicResult_t heuristic_;
-};
-
-} // namespace mlx::core::cu

mlx/backend/cuda/gemms/gemv.cu

@@ -1,173 +0,0 @@
-// Copyright © 2025 Apple Inc.
-
-#include "mlx/backend/cuda/gemms/gemv.h"
-#include "mlx/backend/cuda/kernel_utils.cuh"
-#include "mlx/dtype_utils.h"
-
-#include <cooperative_groups.h>
-#include <cooperative_groups/reduce.h>
-
-namespace mlx::core::cu {
-
-namespace cg = cooperative_groups;
-
-static constexpr int rows_per_block = 8;
-
-template <typename T, int rows_per_block, int n_per_thread>
-__device__ void
-gemv_impl(const T* mat, const T* vec, T* out, int rows, int cols) {
-  auto block = cg::this_thread_block();
-  auto warp = cg::tiled_partition<WARP_SIZE>(block);
-
-  auto g_idx = block.group_index();
-  auto t_idx = block.thread_index();
-  int row = g_idx.x * rows_per_block + t_idx.y;
-
-  if (row < rows) {
-    float sum = 0.0f;
-    for (int col = n_per_thread * warp.thread_rank(); col < cols;
-         col += (WARP_SIZE * n_per_thread)) {
-      auto local_mat =
-          unsafe_load_vector<n_per_thread>(mat + row * cols + col, 0);
-      auto local_vec = unsafe_load_vector<n_per_thread>(vec + col, 0);
-#pragma unroll
-      for (int j = 0; j < n_per_thread; ++j) {
-        sum +=
-            static_cast<float>(local_mat[j]) * static_cast<float>(local_vec[j]);
-      }
-    }
-
-    sum = cg::reduce(warp, sum, cg::plus<float>{});
-    if (warp.thread_rank() == 0) {
-      out[row] = static_cast<T>(sum);
-    }
-  }
-}
-
-template <typename T, int rows_per_block, int n_per_thread>
-__global__ void
-gemv_single(const T* mat, const T* vec, T* out, int rows, int cols) {
-  gemv_impl<T, rows_per_block, n_per_thread>(mat, vec, out, rows, cols);
-}
-
-template <typename T, int rows_per_block, int n_per_thread>
-__global__ void gemv_batched(
-    const T* mat,
-    const T* vec,
-    T* out,
-    int rows,
-    int cols,
-    const __grid_constant__ Shape batch_shape,
-    const __grid_constant__ Strides mat_batch_strides,
-    const __grid_constant__ Strides vec_batch_strides,
-    int batch_ndim) {
-  auto block = cg::this_thread_block();
-  auto batch_idx = block.group_index().y;
-  auto [vec_offset, mat_offset] = elem_to_loc(
-      batch_idx,
-      batch_shape.data(),
-      vec_batch_strides.data(),
-      mat_batch_strides.data(),
-      batch_ndim);
-  gemv_impl<T, rows_per_block, n_per_thread>(
-      mat + mat_offset, vec + vec_offset, out + batch_idx * rows, rows, cols);
-}
-
-bool can_use_gemv(int M, int N, int K, bool a_transposed, bool b_transposed) {
-  return K % 32 == 0 && ((M == 1 && b_transposed) || (N == 1 && !a_transposed));
-}
-
-template <typename F>
-void dispatch_n_per_thread(int n_per_thread, F&& f) {
-  switch (n_per_thread) {
-    case 1:
-      f(std::integral_constant<int, 1>{});
-      break;
-    case 2:
-      f(std::integral_constant<int, 2>{});
-      break;
-    case 4:
-      f(std::integral_constant<int, 4>{});
-      break;
-  }
-}
-
-void gemv(
-    const array& a,
-    const array& b,
-    array& out,
-    int M,
-    int N,
-    int K,
-    uint32_t batch_count,
-    const mlx::core::Shape& batch_shape,
-    const mlx::core::Strides& a_batch_strides,
-    const mlx::core::Strides& b_batch_strides,
-    CommandEncoder& encoder) {
-  encoder.set_input_array(a);
-  encoder.set_input_array(b);
-  encoder.set_output_array(out);
-  dispatch_float_types(out.dtype(), "gemv", [&](auto type_tag) {
-    using DataType = cuda_type_t<MLX_GET_TYPE(type_tag)>;
-    dim3 block_dims{WARP_SIZE, rows_per_block};
-    const DataType* mat;
-    const DataType* vec;
-    int rows;
-    int cols = K;
-    auto mat_strides = const_param(a_batch_strides);
-    auto vec_strides = const_param(b_batch_strides);
-
-    if (M == 1) {
-      mat = b.data<DataType>();
-      vec = a.data<DataType>();
-      rows = N;
-      std::swap(mat_strides, vec_strides);
-    } else {
-      mat = a.data<DataType>();
-      vec = b.data<DataType>();
-      rows = M;
-    }
-    uint32_t num_blocks_x = (rows + rows_per_block - 1) / rows_per_block;
-    int n_per_t;
-    if (K % 128 == 0 && is_aligned<4>(mat) && is_aligned<4>(vec)) {
-      n_per_t = 4;
-    } else if (K % 64 == 0 && is_aligned<2>(mat) && is_aligned<2>(vec)) {
-      n_per_t = 2;
-    } else {
-      n_per_t = 1;
-    }
-    dispatch_n_per_thread(n_per_t, [&](auto n_per_thread) {
-      if (batch_count == 1) {
-        auto kernel = gemv_single<DataType, rows_per_block, n_per_thread()>;
-        encoder.add_kernel_node(
-            kernel,
-            num_blocks_x,
-            block_dims,
-            0,
-            mat,
-            vec,
-            out.data<DataType>(),
-            rows,
-            cols);
-      } else {
-        auto kernel = gemv_batched<DataType, rows_per_block, n_per_thread()>;
-        encoder.add_kernel_node(
-            kernel,
-            dim3{num_blocks_x, batch_count},
-            block_dims,
-            0,
-            mat,
-            vec,
-            out.data<DataType>(),
-            rows,
-            cols,
-            const_param(batch_shape),
-            mat_strides,
-            vec_strides,
-            batch_shape.size());
-      }
-    });
-  });
-}
-
-} // namespace mlx::core::cu

mlx/backend/cuda/gemms/gemv.h

@@ -1,24 +0,0 @@
-// Copyright © 2025 Apple Inc.
-
-#pragma once
-
-#include "mlx/backend/cuda/device.h"
-
-namespace mlx::core::cu {
-
-bool can_use_gemv(int M, int N, int K, bool a_transposed, bool b_transposed);
-
-void gemv(
-    const array& a,
-    const array& b,
-    array& out,
-    int M,
-    int N,
-    int K,
-    uint32_t batch_count,
-    const mlx::core::Shape& batch_shape,
-    const mlx::core::Strides& a_batch_strides,
-    const mlx::core::Strides& b_batch_strides,
-    CommandEncoder& encoder);
-
-} // namespace mlx::core::cu

mlx/backend/cuda/indexing.cpp

@@ -29,12 +29,12 @@ void append_indices_arg(
     const std::vector<array>& inputs,
     int nidx,
     int idx_ndim) {
-  SmallVector<const void*> indices(nidx);
+  std::vector<const void*> indices(nidx);
   for (int i = 0; i < nidx; ++i) {
     indices[i] = inputs[i + 1].data<void>();
   }
   args.append(std::move(indices));
-  SmallVector<int32_t> indices_shape(nidx * idx_ndim);
+  std::vector<int32_t> indices_shape(nidx * idx_ndim);
   for (int i = 0; i < nidx; ++i) {
     std::copy_n(
         inputs[i + 1].shape().begin(),
@@ -42,7 +42,7 @@ void append_indices_arg(
         indices_shape.data() + i * idx_ndim);
   }
   args.append(std::move(indices_shape));
-  SmallVector<int64_t> indices_strides(nidx * idx_ndim);
+  std::vector<int64_t> indices_strides(nidx * idx_ndim);
   for (int i = 0; i < nidx; ++i) {
     std::copy_n(
         inputs[i + 1].strides().begin(),
@@ -110,7 +110,7 @@ void Gather::eval_gpu(const std::vector<array>& inputs, array& out) {
   args.append<int32_t>(src.ndim());
   args.append_ndim(slice_sizes_);
   args.append(slice_size);
-  args.append(SmallVector<int32_t>(axes_.begin(), axes_.end()));
+  args.append(axes_);
   append_indices_arg(args, inputs, nidx, idx_ndim);
 
   std::string kernel_name = fmt::format(
@@ -128,7 +128,7 @@ void Gather::eval_gpu(const std::vector<array>& inputs, array& out) {
   encoder.set_output_array(out);
 
   auto kernel = mod.get_kernel(kernel_name);
-  auto [num_blocks, block_dims] = get_launch_args(out, large);
+  auto [num_blocks, block_dims] = get_launch_args(kernel, out, large);
   encoder.add_kernel_node(kernel, num_blocks, block_dims, 0, args.args());
 }
 
@@ -211,7 +211,7 @@ void Scatter::eval_gpu(const std::vector<array>& inputs, array& out) {
   args.append_ndim(out.shape());
   args.append_ndim(out.strides());
   args.append<int32_t>(out.ndim());
-  args.append(SmallVector<int32_t>(axes_.begin(), axes_.end()));
+  args.append(axes_);
   append_indices_arg(args, inputs, nidx, idx_ndim);
 
   std::string kernel_name = fmt::format(
@@ -229,7 +229,7 @@ void Scatter::eval_gpu(const std::vector<array>& inputs, array& out) {
   }
   encoder.set_output_array(out);
   auto kernel = mod.get_kernel(kernel_name);
-  auto [num_blocks, block_dims] = get_launch_args(upd, large);
+  auto [num_blocks, block_dims] = get_launch_args(kernel, upd, large);
   encoder.add_kernel_node(kernel, num_blocks, block_dims, 0, args.args());
 }
 
@@ -317,7 +317,7 @@ void GatherAxis::eval_gpu(const std::vector<array>& inputs, array& out) {
   }
   encoder.set_output_array(out);
   auto kernel = mod.get_kernel(kernel_name);
-  auto [num_blocks, block_dims] = get_launch_args(idx, large);
+  auto [num_blocks, block_dims] = get_launch_args(kernel, idx, large);
   encoder.add_kernel_node(kernel, num_blocks, block_dims, 0, args.args());
 }
 
@@ -421,7 +421,7 @@ void ScatterAxis::eval_gpu(const std::vector<array>& inputs, array& out) {
   }
   encoder.set_output_array(out);
   auto kernel = mod.get_kernel(kernel_name);
-  auto [num_blocks, block_dims] = get_launch_args(idx, large);
+  auto [num_blocks, block_dims] = get_launch_args(kernel, idx, large);
   encoder.add_kernel_node(kernel, num_blocks, block_dims, 0, args.args());
 }
 

mlx/backend/cuda/iterators/general_iterator.cuh

@@ -0,0 +1,121 @@
+// Copyright © 2025 Apple Inc.
+
+#pragma once
+
+#include <thrust/iterator/iterator_adaptor.h>
+#include <cuda/std/utility>
+
+#include "mlx/backend/cuda/kernel_utils.cuh"
+
+namespace mlx::core::cu {
+
+// Iterating non-contiguous array.
+template <typename Iterator, typename IdxT = int64_t>
+class general_iterator
+    : public thrust::
+          iterator_adaptor<general_iterator<Iterator, IdxT>, Iterator> {
+ public:
+  using super_t =
+      thrust::iterator_adaptor<general_iterator<Iterator, IdxT>, Iterator>;
+
+  using reference = typename super_t::reference;
+  using difference_type = typename super_t::difference_type;
+
+  __host__ __device__ general_iterator(
+      Iterator it,
+      IdxT index,
+      int ndim,
+      Shape shape,
+      Strides strides)
+      : super_t(it),
+        index_(index),
+        ndim_(ndim),
+        shape_(cuda::std::move(shape)),
+        strides_(cuda::std::move(strides)) {}
+
+  __host__ __device__ IdxT index() const {
+    return index_;
+  }
+
+  __host__ __device__ const Shape& shape() const {
+    return shape_;
+  }
+
+  __host__ __device__ const Strides& strides() const {
+    return strides_;
+  }
+
+ private:
+  friend class thrust::iterator_core_access;
+
+  __host__ __device__ bool equal(const general_iterator& other) const {
+    return this->base() == other.base() && this->index() == other.index();
+  }
+
+  __host__ __device__ void advance(difference_type n) {
+    this->index_ += n;
+  }
+
+  __host__ __device__ void increment() {
+    this->index_ += 1;
+  }
+
+  __host__ __device__ void decrement() {
+    this->index_ -= 1;
+  }
+
+  __host__ __device__ difference_type
+  distance_to(const general_iterator& other) const {
+    _CCCL_ASSERT(
+        this->base() == other.base(),
+        "Underlying iterator must point to same base iterator");
+    return other.index() - this->index();
+  }
+
+  // The dereference is device-only to avoid accidental running in host.
+  __device__ typename super_t::reference dereference() const {
+    IdxT offset = elem_to_loc(index_, shape_.data(), strides_.data(), ndim_);
+    return *(this->base() + offset);
+  }
+
+  IdxT index_;
+  int ndim_;
+  Shape shape_;
+  Strides strides_;
+};
+
+template <typename IdxT, typename Iterator>
+__host__ __device__ auto make_general_iterator(
+    Iterator it,
+    IdxT index,
+    int ndim,
+    Shape shape,
+    Strides strides) {
+  return general_iterator<Iterator, IdxT>(
+      it, index, ndim, cuda::std::move(shape), cuda::std::move(strides));
+}
+
+template <typename IdxT, typename Iterator>
+auto make_general_iterator(
+    Iterator it,
+    const std::vector<int32_t>& shape,
+    const std::vector<int64_t>& strides) {
+  return make_general_iterator<IdxT>(
+      it, 0, shape.size(), const_param(shape), const_param(strides));
+}
+
+template <typename IdxT, typename Iterator>
+auto make_general_iterators(
+    Iterator it,
+    IdxT size,
+    const std::vector<int32_t>& shape,
+    const std::vector<int64_t>& strides) {
+  auto ndim = shape.size();
+  auto shape_arg = const_param(shape);
+  auto strides_arg = const_param(strides);
+  return std::make_pair(
+      make_general_iterator<IdxT>(it, 0, ndim, shape_arg, strides_arg),
+      make_general_iterator<IdxT>(it, size, ndim, shape_arg, strides_arg));
+}
+
+} // namespace mlx::core::cu

mlx/backend/cuda/iterators/strided_iterator.cuh

@@ -0,0 +1,60 @@
+// Copyright © 2025 Apple Inc.
+
+#pragma once
+
+#include <thrust/iterator/iterator_adaptor.h>
+#include <thrust/iterator/iterator_facade.h>
+
+namespace mlx::core::cu {
+
+// RandomAccessIterator for strided access to array entries.
+template <typename Iterator, typename Stride = int64_t>
+class strided_iterator
+    : public thrust::
+          iterator_adaptor<strided_iterator<Iterator, Stride>, Iterator> {
+ public:
+  using super_t =
+      thrust::iterator_adaptor<strided_iterator<Iterator, Stride>, Iterator>;
+
+  using reference = typename super_t::reference;
+  using difference_type = typename super_t::difference_type;
+
+  __host__ __device__ strided_iterator(Iterator it, Stride stride)
+      : super_t(it), stride_(stride) {}
+
+  __host__ __device__ Stride stride() const {
+    return stride_;
+  }
+
+ private:
+  friend class thrust::iterator_core_access;
+
+  __host__ __device__ bool equal(const strided_iterator& other) const {
+    return this->base() == other.base();
+  }
+
+  __host__ __device__ void advance(difference_type n) {
+    this->base_reference() += n * stride_;
+  }
+
+  __host__ __device__ void increment() {
+    this->base_reference() += stride_;
+  }
+
+  __host__ __device__ void decrement() {
+    this->base_reference() -= stride_;
+  }
+
+  __host__ __device__ difference_type
+  distance_to(const strided_iterator& other) const {
+    const difference_type dist = other.base() - this->base();
+    _CCCL_ASSERT(
+        dist % stride() == 0,
+        "Underlying iterator difference must be divisible by the stride");
+    return dist / stride();
+  }
+
+  Stride stride_;
+};
+
+} // namespace mlx::core::cu

mlx/backend/cuda/jit_module.cpp

@@ -9,6 +9,7 @@
 #include <cstdlib>
 #include <filesystem>
 #include <fstream>
+#include <unordered_map>
 
 #include <fmt/format.h>
 #include <nvrtc.h>
@@ -329,16 +330,11 @@ CUfunction JitModule::get_kernel(const std::string& kernel_name) {
   return it->second;
 }
 
-std::unordered_map<std::string, JitModule>& get_jit_module_cache() {
-  static std::unordered_map<std::string, JitModule> map;
-  return map;
-}
-
 JitModule& get_jit_module(
     const mlx::core::Device& device,
     const std::string& name,
     const KernelBuilder& builder) {
-  auto& map = get_jit_module_cache();
+  static std::unordered_map<std::string, JitModule> map;
   auto it = map.find(name);
   if (it == map.end()) {
     it = map.try_emplace(name, cu::device(device), name, builder).first;

mlx/backend/cuda/jit_module.h

@@ -40,14 +40,19 @@ struct KernelArgs {
   }
 
   template <typename T>
-  void append(SmallVector<T> vec) {
-    storage_.emplace_back(std::move(vec));
-    append_ptr(std::get<SmallVector<T>>(storage_.back()).data());
+  void append(std::vector<T> vec) {
+    if (vec.empty()) {
+      // The nullptr can not be used as arg, pass something not null.
+      append(std::monostate{});
+    } else {
+      append_ptr(vec.data());
+      storage_.emplace_back(std::move(vec));
+    }
   }
 
   // Make sure the arg is copied to an array with size of NDIM.
   template <size_t NDIM = MAX_NDIM, typename T>
-  void append_ndim(SmallVector<T> vec) {
+  void append_ndim(std::vector<T> vec) {
     if (vec.size() > NDIM) {
       throw std::runtime_error(
           fmt::format("ndim can not be larger than {}.", NDIM));
@@ -71,9 +76,9 @@ struct KernelArgs {
       int32_t,
       uint32_t,
       int64_t,
-      SmallVector<const void*>,
-      SmallVector<int32_t>,
-      SmallVector<int64_t>>;
+      std::vector<const void*>,
+      std::vector<int32_t>,
+      std::vector<int64_t>>;
   std::deque<Arg> storage_;
 };
 
@@ -94,8 +99,6 @@ class JitModule {
   std::unordered_map<std::string, CUfunction> kernels_;
 };
 
-std::unordered_map<std::string, JitModule>& get_jit_module_cache();
-
 JitModule& get_jit_module(
     const mlx::core::Device& device,
     const std::string& name,

mlx/backend/cuda/kernel_utils.cu

@@ -30,25 +30,4 @@ std::pair<dim3, dim3> get_grid_and_block(int dim0, int dim1, int dim2) {
   return std::make_pair(dim3(gx, gy, gz), dim3(bx, by, bz));
 }
 
-std::tuple<dim3, uint> get_launch_args(
-    size_t size,
-    const Shape& shape,
-    const Strides& strides,
-    bool large,
-    int work_per_thread) {
-  size_t nthreads = cuda::ceil_div(size, work_per_thread);
-  uint block_dim = 1024;
-  if (block_dim > nthreads) {
-    block_dim = nthreads;
-  }
-  dim3 num_blocks;
-  if (large) {
-    num_blocks = get_2d_grid_dims(shape, strides, work_per_thread);
-    num_blocks.x = cuda::ceil_div(num_blocks.x, block_dim);
-  } else {
-    num_blocks.x = cuda::ceil_div(nthreads, block_dim);
-  }
-  return std::make_tuple(num_blocks, block_dim);
-}
-
 } // namespace mlx::core

mlx/backend/cuda/kernel_utils.cuh

@@ -101,7 +101,7 @@ inline constexpr bool is_inexact_v = is_floating_v<T> || is_complex_v<T>;
 
 // Utility to copy data from vector to array in host.
 template <int NDIM = MAX_NDIM, typename T = int32_t>
-inline cuda::std::array<T, NDIM> const_param(const SmallVector<T>& vec) {
+inline cuda::std::array<T, NDIM> const_param(const std::vector<T>& vec) {
   if (vec.size() > NDIM) {
     throw std::runtime_error(
         fmt::format("ndim can not be larger than {}.", NDIM));
@@ -120,19 +120,53 @@ dim3 get_2d_grid_dims(
     size_t divisor);
 std::pair<dim3, dim3> get_grid_and_block(int dim0, int dim1, int dim2);
 
+// Return a block size that achieves maximum potential occupancy for kernel.
+template <typename T>
+inline uint max_occupancy_block_dim(T kernel) {
+  int _, block_dim;
+  if constexpr (std::is_same_v<T, CUfunction>) {
+    CHECK_CUDA_ERROR(
+        cuOccupancyMaxPotentialBlockSize(&_, &block_dim, kernel, 0, 0, 0));
+  } else {
+    CHECK_CUDA_ERROR(
+        cudaOccupancyMaxPotentialBlockSize(&_, &block_dim, kernel));
+  }
+  return block_dim;
+}
+
 // Get the num_blocks and block_dims that maximize occupancy for |kernel|,
 // assuming each thread handles |work_per_thread| elements of |arr|.
-std::tuple<dim3, uint> get_launch_args(
+template <typename T>
+inline std::tuple<dim3, uint> get_launch_args(
+    T kernel,
     size_t size,
     const Shape& shape,
     const Strides& strides,
     bool large,
-    int work_per_thread = 1);
+    int work_per_thread = 1) {
+  size_t nthreads = cuda::ceil_div(size, work_per_thread);
+  uint block_dim = max_occupancy_block_dim(kernel);
+  if (block_dim > nthreads) {
+    block_dim = nthreads;
+  }
+  dim3 num_blocks;
+  if (large) {
+    num_blocks = get_2d_grid_dims(shape, strides, work_per_thread);
+    num_blocks.x = cuda::ceil_div(num_blocks.x, block_dim);
+  } else {
+    num_blocks.x = cuda::ceil_div(nthreads, block_dim);
+  }
+  return std::make_tuple(num_blocks, block_dim);
+}
 
-inline std::tuple<dim3, uint>
-get_launch_args(const array& arr, bool large, int work_per_thread = 1) {
+template <typename T>
+inline std::tuple<dim3, uint> get_launch_args(
+    T kernel,
+    const array& arr,
+    bool large,
+    int work_per_thread = 1) {
   return get_launch_args(
-      arr.size(), arr.shape(), arr.strides(), large, work_per_thread);
+      kernel, arr.size(), arr.shape(), arr.strides(), large, work_per_thread);
 }
 
 } // namespace mlx::core

mlx/backend/cuda/layer_norm.cu

@@ -1,6 +1,7 @@
 // Copyright © 2025 Apple Inc.
 
 #include "mlx/backend/cuda/device.h"
+#include "mlx/backend/cuda/iterators/strided_iterator.cuh"
 #include "mlx/backend/cuda/kernel_utils.cuh"
 #include "mlx/backend/cuda/reduce/reduce.cuh"
 #include "mlx/backend/gpu/copy.h"
@@ -10,6 +11,8 @@
 #include <cooperative_groups.h>
 #include <cooperative_groups/reduce.h>
 #include <nvtx3/nvtx3.hpp>
+#include <cub/block/block_load.cuh>
+#include <cub/block/block_reduce.cuh>
 
 namespace mlx::core {
 
@@ -72,11 +75,9 @@ __global__ void layer_norm(
   float sum = 0;
   for (int r = 0; r < cuda::ceil_div(axis_size, BLOCK_DIM * N_READS); ++r) {
     auto index = r * BLOCK_DIM + block.thread_rank();
-    auto xn = load_vector<N_READS>(x, index, axis_size, T(0));
-#pragma unroll
-    for (int i = 0; i < N_READS; ++i) {
-      sum += static_cast<float>(xn[i]);
-    }
+    T xn[N_READS] = {};
+    cub::LoadDirectBlocked(index, x, xn, axis_size);
+    sum += static_cast<float>(cub::ThreadReduce(xn, cuda::std::plus<>{}));
   }
   sum = BlockReduceT{block, temp}.Sum(sum);
 
@@ -87,18 +88,11 @@ __global__ void layer_norm(
   float normalizer = 0;
   for (int r = 0; r < cuda::ceil_div(axis_size, BLOCK_DIM * N_READS); ++r) {
     auto index = r * BLOCK_DIM + block.thread_rank();
-    if ((index + 1) * N_READS <= axis_size) {
-      auto xn = load_vector<N_READS>(x, index);
-#pragma unroll
-      for (int i = 0; i < N_READS; ++i) {
-        float t = static_cast<float>(xn[i]) - mean;
-        normalizer += t * t;
-      }
-    } else {
-      for (int i = index * N_READS; i < axis_size; ++i) {
-        float t = static_cast<float>(x[i]) - mean;
-        normalizer += t * t;
-      }
+    T xn[N_READS];
+    cub::LoadDirectBlocked(index, x, xn, axis_size, mean);
+    for (int i = 0; i < N_READS; ++i) {
+      float t = static_cast<float>(xn[i]) - mean;
+      normalizer += t * t;
     }
   }
   normalizer = BlockReduceT{block, temp}.Sum(normalizer);
@@ -107,15 +101,17 @@ __global__ void layer_norm(
   // Outputs.
   for (int r = 0; r < cuda::ceil_div(axis_size, BLOCK_DIM * N_READS); ++r) {
     auto index = r * BLOCK_DIM + block.thread_rank();
-    auto xn = load_vector<N_READS>(x, index, axis_size, T(0));
-    auto wn = load_vector<N_READS>(w, index, axis_size, w_stride, T(0));
-    auto bn = load_vector<N_READS>(b, index, axis_size, b_stride, T(0));
-#pragma unroll
+    T xn[N_READS];
+    T wn[N_READS];
+    T bn[N_READS];
+    cub::LoadDirectBlocked(index, x, xn, axis_size);
+    cub::LoadDirectBlocked(index, strided_iterator(w, w_stride), wn, axis_size);
+    cub::LoadDirectBlocked(index, strided_iterator(b, b_stride), bn, axis_size);
     for (int i = 0; i < N_READS; ++i) {
       float norm = (static_cast<float>(xn[i]) - mean) * normalizer;
       xn[i] = wn[i] * static_cast<T>(norm) + bn[i];
     }
-    store_vector<N_READS>(out, index, xn, axis_size);
+    cub::StoreDirectBlocked(index, out, xn, axis_size);
   }
 }
 
@@ -148,11 +144,9 @@ __global__ void layer_norm_vjp(
   float sum = 0;
   for (int r = 0; r < cuda::ceil_div(axis_size, BLOCK_DIM * N_READS); ++r) {
     auto index = r * BLOCK_DIM + block.thread_rank();
-    auto xn = load_vector<N_READS>(x, index, axis_size, T(0));
-#pragma unroll
-    for (int i = 0; i < N_READS; ++i) {
-      sum += static_cast<float>(xn[i]);
-    }
+    T xn[N_READS] = {};
+    cub::LoadDirectBlocked(index, x, xn, axis_size);
+    sum += static_cast<float>(cub::ThreadReduce(xn, cuda::std::plus<>{}));
   }
   sum = BlockReduceF{block, temp.f}.Sum(sum);
 
@@ -162,28 +156,19 @@ __global__ void layer_norm_vjp(
   // Normalizer.
   float3 factors = {};
   for (int r = 0; r < cuda::ceil_div(axis_size, BLOCK_DIM * N_READS); ++r) {
+    T xn[N_READS];
+    T wn[N_READS] = {};
+    T gn[N_READS] = {};
     auto index = r * BLOCK_DIM + block.thread_rank();
-    auto gn = load_vector<N_READS>(g, index, axis_size, T(0));
-    auto wn = load_vector<N_READS>(w, index, axis_size, w_stride, T(0));
-
-    if ((index + 1) * N_READS <= axis_size) {
-      auto xn = load_vector<N_READS>(x, index);
-#pragma unroll
-      for (int i = 0; i < N_READS; ++i) {
-        float t = static_cast<float>(xn[i]) - mean;
-        float wi = wn[i];
-        float gi = gn[i];
-        float wg = wi * gi;
-        factors = plus_f3(factors, {wg, wg * t, t * t});
-      }
-    } else {
-      for (int i = index * N_READS; i < axis_size; ++i) {
-        float t = static_cast<float>(x[i]) - mean;
-        float wi = wn[i];
-        float gi = gn[i];
-        float wg = wi * gi;
-        factors = plus_f3(factors, {wg, wg * t, t * t});
-      }
+    cub::LoadDirectBlocked(index, x, xn, axis_size, mean);
+    cub::LoadDirectBlocked(index, g, gn, axis_size);
+    cub::LoadDirectBlocked(index, strided_iterator(w, w_stride), wn, axis_size);
+    for (int i = 0; i < N_READS; i++) {
+      float t = static_cast<float>(xn[i]) - mean;
+      float wi = wn[i];
+      float gi = gn[i];
+      float wg = wi * gi;
+      factors = plus_f3(factors, {wg, wg * t, t * t});
     }
   }
   factors = BlockReduceF3{block, temp.f3}.Reduce(factors, plus_f3, {});
@@ -195,10 +180,12 @@ __global__ void layer_norm_vjp(
   // Outputs.
   for (int r = 0; r < cuda::ceil_div(axis_size, BLOCK_DIM * N_READS); ++r) {
     auto index = r * BLOCK_DIM + block.thread_rank();
-    auto xn = load_vector<N_READS>(x, index, axis_size, T(0));
-    auto gn = load_vector<N_READS>(g, index, axis_size, T(0));
-    auto wn = load_vector<N_READS>(w, index, axis_size, w_stride, T(0));
-
+    T xn[N_READS];
+    T wn[N_READS];
+    T gn[N_READS];
+    cub::LoadDirectBlocked(index, x, xn, axis_size);
+    cub::LoadDirectBlocked(index, g, gn, axis_size);
+    cub::LoadDirectBlocked(index, strided_iterator(w, w_stride), wn, axis_size);
     for (int i = 0; i < N_READS; i++) {
       float xi = (static_cast<float>(xn[i]) - mean) * normalizer;
       float wi = wn[i];
@@ -208,9 +195,9 @@ __global__ void layer_norm_vjp(
         wn[i] = gi * xi;
       }
     }
-    store_vector<N_READS>(gx, index, xn, axis_size);
+    cub::StoreDirectBlocked(index, gx, xn, axis_size);
     if constexpr (HAS_W) {
-      store_vector<N_READS>(gw, index, wn, axis_size);
+      cub::StoreDirectBlocked(index, gw, wn, axis_size);
     }
   }
 }
@@ -271,9 +258,9 @@ void LayerNorm::eval_gpu(
   encoder.set_input_array(b);
   encoder.set_output_array(out);
   dispatch_float_types(out.dtype(), "layernorm", [&](auto type_tag) {
-    using DataType = cuda_type_t<MLX_GET_TYPE(type_tag)>;
-    constexpr int N_READS = 16 / sizeof(DataType);
+    constexpr uint32_t N_READS = 4;
     dispatch_block_dim(cuda::ceil_div(axis_size, N_READS), [&](auto block_dim) {
+      using DataType = cuda_type_t<MLX_GET_TYPE(type_tag)>;
       auto kernel = cu::layer_norm<DataType, block_dim(), N_READS>;
       encoder.add_kernel_node(
           kernel,
@@ -379,10 +366,10 @@ void LayerNormVJP::eval_gpu(
   encoder.set_output_array(gw_temp);
   dispatch_float_types(gx.dtype(), "layernorm_vjp", [&](auto type_tag) {
     dispatch_bool(has_w, [&](auto has_w_constant) {
-      using DataType = cuda_type_t<MLX_GET_TYPE(type_tag)>;
-      constexpr int N_READS = 16 / sizeof(DataType);
+      constexpr int N_READS = 4;
       dispatch_block_dim(
           cuda::ceil_div(axis_size, N_READS), [&](auto block_dim) {
+            using DataType = cuda_type_t<MLX_GET_TYPE(type_tag)>;
             auto kernel = cu::layer_norm_vjp<
                 DataType,
                 has_w_constant.value,

mlx/backend/cuda/logsumexp.cu

@@ -43,19 +43,20 @@ __global__ void logsumexp(const T* in, T* out, int axis_size) {
   AccT maxval = Limits<AccT>::finite_min();
   AccT normalizer = 0;
   for (int r = 0; r < cuda::ceil_div(axis_size, BLOCK_DIM * N_READS); r++) {
-    auto index = r * BLOCK_DIM + block.thread_rank();
-    auto vals = load_vector<N_READS>(in, index, axis_size, Limits<T>::min());
+    AccT vals[N_READS];
+    cub::LoadDirectBlocked(
+        r * BLOCK_DIM + block.thread_rank(),
+        make_cast_iterator<AccT>(in),
+        vals,
+        axis_size,
+        Limits<AccT>::min());
     prevmax = maxval;
-#pragma unroll
-    for (int i = 0; i < N_READS; ++i) {
-      maxval = max_op(maxval, static_cast<AccT>(vals[i]));
-    }
+    maxval = max_op(maxval, cub::ThreadReduce(vals, max_op));
     // Online normalizer calculation for softmax:
     // https://github.com/NVIDIA/online-softmax
     normalizer = normalizer * softmax_exp(prevmax - maxval);
     for (int i = 0; i < N_READS; i++) {
-      normalizer =
-          normalizer + softmax_exp(static_cast<AccT>(vals[i]) - maxval);
+      normalizer = normalizer + softmax_exp(vals[i] - maxval);
     }
   }
 
@@ -142,9 +143,9 @@ void LogSumExp::eval_gpu(const std::vector<array>& inputs, array& out) {
   encoder.set_input_array(in);
   encoder.set_output_array(out);
   dispatch_float_types(out.dtype(), "logsumexp", [&](auto type_tag) {
-    using DataType = cuda_type_t<MLX_GET_TYPE(type_tag)>;
-    constexpr int N_READS = 16 / sizeof(DataType);
+    constexpr int N_READS = 4;
     dispatch_block_dim(cuda::ceil_div(axis_size, N_READS), [&](auto block_dim) {
+      using DataType = cuda_type_t<MLX_GET_TYPE(type_tag)>;
       auto kernel = cu::logsumexp<DataType, float, block_dim(), N_READS>;
       encoder.add_kernel_node(
           kernel,

mlx/backend/cuda/lru_cache.h

@@ -1,159 +0,0 @@
-// Copyright © 2025 Apple Inc.
-
-#pragma once
-
-#include <cstring>
-#include <list>
-#include <unordered_map>
-#include <utility>
-
-namespace mlx::core {
-
-template <
-    typename K,
-    typename V,
-    template <typename...> typename M = std::unordered_map>
-class LRUCache {
- public:
-  using value_type = std::pair<K, V>;
-  using list_type = std::list<value_type>;
-  using iterator = typename list_type::iterator;
-  using const_iterator = typename list_type::const_iterator;
-  using map_type = M<K, iterator>;
-
-  explicit LRUCache(size_t capacity) : capacity_(capacity) {
-    if (capacity == 0) {
-      throw std::runtime_error("LRUCache requires capacity > 0.");
-    }
-  }
-
-  size_t size() const {
-    return map_.size();
-  }
-  size_t capacity() const {
-    return capacity_;
-  }
-  bool empty() const {
-    return vlist_.empty();
-  }
-
-  void resize(size_t new_capacity) {
-    capacity_ = new_capacity;
-    trim();
-  }
-
-  iterator begin() {
-    return vlist_.begin();
-  }
-  const_iterator begin() const {
-    return vlist_.begin();
-  }
-  iterator end() {
-    return vlist_.end();
-  }
-  const_iterator end() const {
-    return vlist_.end();
-  }
-
-  void clear() {
-    map_.clear();
-    vlist_.clear();
-  }
-
-  iterator find(const K& key) {
-    auto it = map_.find(key);
-    if (it == map_.end())
-      return end();
-    vlist_.splice(vlist_.begin(), vlist_, it->second);
-    return it->second;
-  }
-
-  template <typename U>
-  std::pair<iterator, bool> emplace(const K& key, U&& value) {
-    auto it = map_.find(key);
-    if (it != map_.end()) {
-      vlist_.splice(vlist_.begin(), vlist_, it->second);
-      return {it->second, false};
-    }
-
-    vlist_.emplace_front(key, std::forward<U>(value));
-    map_[key] = vlist_.begin();
-
-    trim();
-
-    return {vlist_.begin(), true};
-  }
-
-  iterator erase(iterator pos) {
-    map_.erase(pos->first);
-    return vlist_.erase(pos);
-  }
-
-  V& operator[](const K& key) {
-    auto it = find(key);
-    if (it == end()) {
-      it = emplace(key, V{}).first;
-    }
-    return it->second;
-  }
-
- private:
-  void trim() {
-    while (map_.size() > capacity_) {
-      auto last = std::prev(vlist_.end());
-      map_.erase(last->first);
-      vlist_.pop_back();
-    }
-  }
-
-  list_type vlist_;
-  map_type map_;
-  size_t capacity_;
-};
-
-// Turn a POD struct into a container key by doing bytes compare.
-template <typename T>
-struct BytesKey {
-  T pod;
-  static_assert(std::is_standard_layout_v<T>, "T is not POD");
-
-  BytesKey(T pod) : pod(std::move(pod)) {}
-
-  BytesKey(const BytesKey& other) {
-    memcpy(&pod, &other.pod, sizeof(T));
-  }
-
-  BytesKey(BytesKey&& other) {
-    memcpy(&pod, &other.pod, sizeof(T));
-  }
-
-  bool operator==(const BytesKey& other) const {
-    auto* ptr1 = reinterpret_cast<const uint8_t*>(&pod);
-    auto* ptr2 = reinterpret_cast<const uint8_t*>(&other.pod);
-    return memcmp(ptr1, ptr2, sizeof(T)) == 0;
-  }
-};
-
-// Compute hash according to the bytes value of T.
-template <typename T>
-struct BytesHash {
-  static_assert(std::is_standard_layout_v<T>, "T is not POD");
-
-  size_t operator()(const T& pod) const {
-    auto* ptr = reinterpret_cast<const uint8_t*>(&pod);
-    uint32_t value = 0x811C9DC5;
-    for (int i = 0; i < sizeof(T); ++i) {
-      value ^= ptr[i];
-      value *= 0x01000193;
-    }
-    return value;
-  }
-};
-
-template <typename K, typename V>
-using BytesKeyHashMap = std::unordered_map<K, V, BytesHash<K>>;
-
-template <typename K, typename V>
-using LRUBytesKeyCache = LRUCache<BytesKey<K>, V, BytesKeyHashMap>;
-
-} // namespace mlx::core

mlx/backend/cuda/matmul.cpp

@@ -2,15 +2,289 @@
 
 #include "mlx/backend/common/matmul.h"
 #include "mlx/backend/cuda/device.h"
-#include "mlx/backend/cuda/gemms/cublas_gemm.h"
-#include "mlx/backend/cuda/gemms/gemv.h"
 #include "mlx/backend/gpu/copy.h"
+#include "mlx/dtype_utils.h"
 #include "mlx/primitives.h"
+#include "mlx/utils.h"
 
+#include <cublasLt.h>
+#include <fmt/format.h>
 #include <nvtx3/nvtx3.hpp>
+
 #include <numeric>
 
 namespace mlx::core {
+
+namespace cu {
+
+#define CHECK_CUBLAS_ERROR(cmd) check_cublas_error(#cmd, (cmd))
+
+void check_cublas_error(const char* name, cublasStatus_t err) {
+  if (err != CUBLAS_STATUS_SUCCESS) {
+    // TODO: Use cublasGetStatusString when it is widely available.
+    throw std::runtime_error(
+        fmt::format("{} failed with code: {}.", name, static_cast<int>(err)));
+  }
+}
+
+struct CublasPreference {
+  CublasPreference(Device& device) {
+    // The recommended cublas workspace size is 4 MiB for pre-Hopper and 32 MiB
+    // for Hopper+:
+    // https://docs.nvidia.com/cuda/cublas/#cublassetworkspace
+    uint64_t MiB = 1024 * 1024;
+    uint64_t workspace_size =
+        device.compute_capability_major() >= 9 ? 32 * MiB : 4 * MiB;
+
+    CHECK_CUBLAS_ERROR(cublasLtMatmulPreferenceCreate(&pref_));
+    CHECK_CUBLAS_ERROR(cublasLtMatmulPreferenceSetAttribute(
+        pref_,
+        CUBLASLT_MATMUL_PREF_MAX_WORKSPACE_BYTES,
+        &workspace_size,
+        sizeof(uint64_t)));
+  }
+
+  ~CublasPreference() {
+    CHECK_CUBLAS_ERROR(cublasLtMatmulPreferenceDestroy(pref_));
+  }
+
+  cublasLtMatmulPreference_t pref_{nullptr};
+};
+
+cublasLtMatmulPreference_t cublas_preference(Device& device) {
+  static CublasPreference pref(device);
+  return pref.pref_;
+}
+
+class MatMul {
+ public:
+  MatMul(
+      Device& device,
+      Dtype dtype,
+      bool a_transposed,
+      uint64_t a_rows,
+      uint64_t a_cols,
+      int64_t lda,
+      bool b_transposed,
+      uint64_t b_rows,
+      uint64_t b_cols,
+      int64_t ldb,
+      int32_t batch_count,
+      int64_t a_batch_stride,
+      int64_t b_batch_stride)
+      : handle_(device.lt_handle()), pref_(cublas_preference(device)) {
+    heuristic_.state = CUBLAS_STATUS_NOT_INITIALIZED;
+
+    auto scale_type = dtype_to_cuda_type(dtype);
+    if (dtype == bfloat16 || dtype == float16) {
+      scale_type = CUDA_R_32F;
+    }
+    CHECK_CUBLAS_ERROR(cublasLtMatmulDescCreate(
+        &matmul_desc_, dtype_to_compute_type(dtype), scale_type));
+    int32_t pointer_mode = CUBLASLT_POINTER_MODE_HOST;
+    CHECK_CUBLAS_ERROR(cublasLtMatmulDescSetAttribute(
+        matmul_desc_,
+        CUBLASLT_MATMUL_DESC_POINTER_MODE,
+        &pointer_mode,
+        sizeof(int32_t)));
+    cublasOperation_t op = CUBLAS_OP_N;
+    CHECK_CUBLAS_ERROR(cublasLtMatmulDescSetAttribute(
+        matmul_desc_,
+        CUBLASLT_MATMUL_DESC_TRANSA,
+        &op,
+        sizeof(cublasOperation_t)));
+    CHECK_CUBLAS_ERROR(cublasLtMatmulDescSetAttribute(
+        matmul_desc_,
+        CUBLASLT_MATMUL_DESC_TRANSB,
+        &op,
+        sizeof(cublasOperation_t)));
+
+    auto type = dtype_to_cuda_type(dtype);
+    a_desc_ = create_matrix_layout(
+        type, a_rows, a_cols, a_transposed, lda, batch_count, a_batch_stride);
+    b_desc_ = create_matrix_layout(
+        type, b_rows, b_cols, b_transposed, ldb, batch_count, b_batch_stride);
+    out_desc_ = create_matrix_layout(
+        type, a_rows, b_cols, false, b_cols, batch_count, a_rows * b_cols);
+  }
+
+  MatMul(
+      Device& device,
+      Dtype dtype,
+      bool a_transposed,
+      uint64_t a_rows,
+      uint64_t a_cols,
+      int64_t lda,
+      bool b_transposed,
+      uint64_t b_rows,
+      uint64_t b_cols,
+      int64_t ldb,
+      int64_t ldc,
+      int32_t batch_count,
+      int64_t a_batch_stride,
+      int64_t b_batch_stride,
+      int64_t c_batch_stride)
+      : MatMul(
+            device,
+            dtype,
+            a_transposed,
+            a_rows,
+            a_cols,
+            lda,
+            b_transposed,
+            b_rows,
+            b_cols,
+            ldb,
+            batch_count,
+            a_batch_stride,
+            b_batch_stride) {
+    auto type = dtype_to_cuda_type(dtype);
+    c_desc_ = create_matrix_layout(
+        type, a_rows, b_cols, false, ldc, batch_count, c_batch_stride);
+  }
+
+  ~MatMul() {
+    CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutDestroy(a_desc_));
+    CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutDestroy(b_desc_));
+    CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutDestroy(c_desc_));
+    CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutDestroy(out_desc_));
+    CHECK_CUBLAS_ERROR(cublasLtMatmulDescDestroy(matmul_desc_));
+  }
+
+  void run(
+      cu::CommandEncoder& encoder,
+      void* out,
+      void* a,
+      void* b,
+      void* c = nullptr,
+      float alpha = 1,
+      float beta = 0) {
+    if (heuristic_.state != CUBLAS_STATUS_SUCCESS) {
+      int ret = 0;
+      CHECK_CUBLAS_ERROR(cublasLtMatmulAlgoGetHeuristic(
+          handle_,
+          matmul_desc_,
+          a_desc_,
+          b_desc_,
+          out_desc_,
+          out_desc_,
+          pref_,
+          1,
+          &heuristic_,
+          &ret));
+      if (ret == 0) {
+        throw std::runtime_error("Can not find algorithm for matmul.");
+      }
+    }
+
+    void* workspace_ptr = nullptr;
+    if (heuristic_.workspaceSize > 0) {
+      array workspace(
+          allocator::malloc(heuristic_.workspaceSize),
+          {static_cast<int>(heuristic_.workspaceSize)},
+          int8);
+      encoder.add_temporary(workspace);
+      workspace_ptr = workspace.data<void>();
+    }
+
+    auto capture = encoder.capture_context();
+    CHECK_CUBLAS_ERROR(cublasLtMatmul(
+        handle_,
+        matmul_desc_,
+        &alpha,
+        a,
+        a_desc_,
+        b,
+        b_desc_,
+        &beta,
+        c ? c : out,
+        c ? c_desc_ : out_desc_,
+        out,
+        out_desc_,
+        &heuristic_.algo,
+        workspace_ptr,
+        heuristic_.workspaceSize,
+        encoder.stream()));
+  }
+
+ private:
+  cublasComputeType_t dtype_to_compute_type(Dtype dtype) {
+    switch (dtype) {
+      case float16:
+        return CUBLAS_COMPUTE_32F;
+      case bfloat16:
+        return CUBLAS_COMPUTE_32F;
+      case float32:
+        return mlx::core::env::enable_tf32() ? CUBLAS_COMPUTE_32F_FAST_TF32
+                                             : CUBLAS_COMPUTE_32F;
+      case float64:
+      case complex64:
+        return CUBLAS_COMPUTE_64F;
+      default:
+        throw std::runtime_error(fmt::format(
+            "Unsupported dtype in MatMul: {}.", dtype_to_string(dtype)));
+    }
+  }
+
+  cudaDataType_t dtype_to_cuda_type(Dtype dtype) {
+    switch (dtype) {
+      case float16:
+        return CUDA_R_16F;
+      case bfloat16:
+        return CUDA_R_16BF;
+      case float32:
+        return CUDA_R_32F;
+      case float64:
+        return CUDA_R_64F;
+      case complex64:
+        return CUDA_C_32F;
+      default:
+        throw std::runtime_error(fmt::format(
+            "Unsupported dtype in MatMul: {}.", dtype_to_string(dtype)));
+    }
+  }
+
+  cublasLtMatrixLayout_t create_matrix_layout(
+      cudaDataType_t type,
+      uint64_t rows,
+      uint64_t cols,
+      bool transposed,
+      int64_t ld,
+      int32_t batch_count,
+      int64_t batch_stride) {
+    cublasLtMatrixLayout_t desc;
+    CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutCreate(&desc, type, rows, cols, ld));
+    cublasLtOrder_t order =
+        transposed ? CUBLASLT_ORDER_COL : CUBLASLT_ORDER_ROW;
+    CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutSetAttribute(
+        desc, CUBLASLT_MATRIX_LAYOUT_ORDER, &order, sizeof(cublasLtOrder_t)));
+    if (batch_count > 1) {
+      CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutSetAttribute(
+          desc,
+          CUBLASLT_MATRIX_LAYOUT_BATCH_COUNT,
+          &batch_count,
+          sizeof(int32_t)));
+      CHECK_CUBLAS_ERROR(cublasLtMatrixLayoutSetAttribute(
+          desc,
+          CUBLASLT_MATRIX_LAYOUT_STRIDED_BATCH_OFFSET,
+          &batch_stride,
+          sizeof(int64_t)));
+    }
+    return desc;
+  }
+
+  cublasLtMatmulPreference_t pref_{nullptr};
+  cublasLtHandle_t handle_{nullptr};
+  cublasLtMatmulDesc_t matmul_desc_{nullptr};
+  cublasLtMatrixLayout_t a_desc_{nullptr};
+  cublasLtMatrixLayout_t b_desc_{nullptr};
+  cublasLtMatrixLayout_t c_desc_{nullptr};
+  cublasLtMatrixLayout_t out_desc_{nullptr};
+  cublasLtMatmulHeuristicResult_t heuristic_;
+};
+
+} // namespace cu
+
 namespace {
 
 std::tuple<bool, int64_t, array>
@@ -79,25 +353,10 @@ void Matmul::eval_gpu(const std::vector<array>& inputs, array& out) {
     batch_shape = {1};
   }
 
-  if (cu::can_use_gemv(M, N, K, a_transposed, b_transposed)) {
-    cu::gemv(
-        a,
-        b,
-        out,
-        M,
-        N,
-        K,
-        batch_count,
-        batch_shape,
-        a_batch_strides,
-        b_batch_strides,
-        encoder);
-    return;
-  }
-
   /////////////////////////////////////////////////////////////////////////////
   // Invoke cublasLt
-  cu::Matmul matmul(
+
+  cu::MatMul matmul(
       cu::device(s.device),
       a.dtype(),
       a_transposed,
@@ -112,13 +371,27 @@ void Matmul::eval_gpu(const std::vector<array>& inputs, array& out) {
       a_batch_strides.back(),
       b_batch_strides.back());
 
-  if ((batch_count / batch_shape.back()) == 1) {
-    matmul.run(encoder, out, a, b);
+  encoder.set_input_array(a);
+  encoder.set_input_array(b);
+  encoder.set_output_array(out);
+  auto nbatch = batch_count / batch_shape.back();
+  if (nbatch == 1) {
+    matmul.run(encoder, out.data<int8_t>(), a.data<int8_t>(), b.data<int8_t>());
     return;
   }
 
-  matmul.run_batched(
-      encoder, out, a, b, batch_shape, a_batch_strides, b_batch_strides);
+  ContiguousIterator a_it(batch_shape, a_batch_strides, batch_shape.size() - 1);
+  ContiguousIterator b_it(batch_shape, b_batch_strides, batch_shape.size() - 1);
+  auto concurrent = encoder.concurrent_context();
+  for (size_t i = 0; i < nbatch; ++i) {
+    matmul.run(
+        encoder,
+        out.data<int8_t>() + out.itemsize() * i * batch_shape.back() * M * N,
+        a.data<int8_t>() + a.itemsize() * a_it.loc,
+        b.data<int8_t>() + b.itemsize() * b_it.loc);
+    a_it.step();
+    b_it.step();
+  }
 }
 
 void AddMM::eval_gpu(const std::vector<array>& inputs, array& out) {
@@ -186,7 +459,7 @@ void AddMM::eval_gpu(const std::vector<array>& inputs, array& out) {
   /////////////////////////////////////////////////////////////////////////////
   // Invoke cublasLt
 
-  cu::Matmul matmul(
+  cu::MatMul matmul(
       cu::device(s.device),
       a.dtype(),
       a_transposed,
@@ -203,22 +476,41 @@ void AddMM::eval_gpu(const std::vector<array>& inputs, array& out) {
       b_batch_strides.back(),
       c_batch_strides.back());
 
-  if ((batch_count / batch_shape.back()) == 1) {
-    matmul.run(encoder, out, a, b, c, alpha_, beta_);
+  encoder.set_input_array(a);
+  encoder.set_input_array(b);
+  encoder.set_input_array(c);
+  encoder.set_output_array(out);
+
+  auto nbatch = batch_count / batch_shape.back();
+  if (nbatch == 1) {
+    matmul.run(
+        encoder,
+        out.data<int8_t>(),
+        a.data<int8_t>(),
+        b.data<int8_t>(),
+        c.data<int8_t>(),
+        alpha_,
+        beta_);
     return;
   }
-  matmul.run_batched(
-      encoder,
-      out,
-      a,
-      b,
-      c,
-      batch_shape,
-      a_batch_strides,
-      b_batch_strides,
-      c_batch_strides,
-      alpha_,
-      beta_);
+
+  ContiguousIterator a_it(batch_shape, a_batch_strides, batch_shape.size() - 1);
+  ContiguousIterator b_it(batch_shape, b_batch_strides, batch_shape.size() - 1);
+  ContiguousIterator c_it(batch_shape, c_batch_strides, batch_shape.size() - 1);
+  auto concurrent = encoder.concurrent_context();
+  for (size_t i = 0; i < nbatch; ++i) {
+    matmul.run(
+        encoder,
+        out.data<int8_t>() + out.itemsize() * i * batch_shape.back() * M * N,
+        a.data<int8_t>() + a.itemsize() * a_it.loc,
+        b.data<int8_t>() + b.itemsize() * b_it.loc,
+        c.data<int8_t>() + c.itemsize() * c_it.loc,
+        alpha_,
+        beta_);
+    a_it.step();
+    b_it.step();
+    c_it.step();
+  }
 }
 
 } // namespace mlx::core

mlx/backend/cuda/matmul/mma.cuh

@@ -2,17 +2,10 @@
 
 #pragma once
 
-#include "mlx/backend/cuda/steel/defines.cuh"
-#include "mlx/backend/cuda/steel/tiles.cuh"
+#include "mlx/backend/cuda/matmul/tiles.cuh"
 
 namespace mlx::core::cu {
 
-/**
- * Fallback mma.
- *
- * We should probably a) implement a fallback or complain about it to the
- * compiler.
- */
 template <typename U, typename T>
 __device__ inline void
 mma_t(Tile16x16<U>& C, Tile16x16<T>& A, Tile16x16<T>& B) {}
@@ -23,11 +16,10 @@ mma_t(Tile16x16<U>& C, Tile16x16<T>& A, Tile16x16<T>& B) {}
  *
  * We actually perform C += A @ B.T
  */
-__device__ __forceinline__ void mma_t(
+__device__ inline void mma_t(
     Tile16x16<float>& C,
     Tile16x16<__nv_bfloat16>& A,
     Tile16x16<__nv_bfloat16>& B) {
-#if defined(MLX_CUDA_SM_80_ENABLED)
   asm volatile(
       "mma.sync.aligned.m16n8k16.row.col.f32.bf16.bf16.f32 "
       "{%0, %1, %2, %3}, "
@@ -84,14 +76,13 @@ __device__ __forceinline__ void mma_t(
         "f"(C.values[2].y),
         "f"(C.values[3].x),
         "f"(C.values[3].y));
-#endif
 }
 
 /**
  * Multiply larger register tiles by delegating to mma_t.
  */
 template <typename U, typename T, int M, int N, int K>
-__device__ __forceinline__ void mma_t(
+__device__ inline void mma_t(
     RegisterTile<U, M, N>& C,
     RegisterTile<T, M, K>& A,
     RegisterTile<T, N, K>& B) {

mlx/backend/cuda/matmul/tiles.cuh

@@ -2,7 +2,7 @@
 
 #pragma once
 
-#include "mlx/backend/cuda/steel/utils.cuh"
+#define MLX_UNROLL _Pragma("unroll")
 
 namespace mlx::core::cu {
 
@@ -67,7 +67,7 @@ struct Tile16x16 {
    * See
    * https://docs.nvidia.com/cuda/parallel-thread-execution/#warp-level-matrix-instructions-ldmatrix
    */
-  __device__ __forceinline__ void load(uint32_t row_address) {
+  __device__ inline void load(uint32_t row_address) {
     if constexpr (
         std::is_same_v<T2, __nv_bfloat162> || std::is_same_v<T2, __half2>) {
       asm volatile(
@@ -169,7 +169,7 @@ struct RegisterTile {
   }
 
   template <typename Tile>
-  __device__ __forceinline__ void
+  __device__ inline void
   load(Tile& tile, uint32_t base_address, int row, int col) {
     MLX_UNROLL
     for (int i = 0; i < TILES_Y; i++) {
@@ -181,22 +181,6 @@ struct RegisterTile {
     }
   }
 
-  template <typename Tile, typename F>
-  __device__ __forceinline__ void
-  load(Tile& tile, F f, uint32_t base_address, int row, int col) {
-    MLX_UNROLL
-    for (int i = 0; i < TILES_Y; i++) {
-      MLX_UNROLL
-      for (int j = 0; j < TILES_X; j++) {
-        f(data[i * TILES_X + j],
-          tile,
-          base_address,
-          row + i * 16,
-          col + j * 16);
-      }
-    }
-  }
-
   template <typename U>
   __device__ inline void store_global(U* x, int N, int row, int col) {
     MLX_UNROLL
@@ -223,57 +207,6 @@ struct RegisterTile {
   }
 };
 
-/**
- * A simple container of multiple Tile16x16.
- *
- * Provides utility functions for loading and manipulating collections of basic
- * tiles.
- */
-template <typename T, int ROWS_, int COLS_>
-struct RegisterTile {
-  static constexpr int ROWS = ROWS_;
-  static constexpr int COLS = COLS_;
-  static constexpr int TILES_X = COLS / 16;
-  static constexpr int TILES_Y = ROWS / 16;
-
-  Tile16x16<T> data[TILES_X * TILES_Y];
-
-  __device__ inline void fill(T v) {
-    MLX_UNROLL
-    for (int i = 0; i < TILES_Y; i++) {
-      MLX_UNROLL
-      for (int j = 0; j < TILES_X; j++) {
-        data[i * TILES_X + j].fill(v);
-      }
-    }
-  }
-
-  template <typename Tile>
-  __device__ inline void
-  load(Tile& tile, uint32_t base_address, int row, int col) {
-    MLX_UNROLL
-    for (int i = 0; i < TILES_Y; i++) {
-      MLX_UNROLL
-      for (int j = 0; j < TILES_X; j++) {
-        data[i * TILES_X + j].load(
-            tile.loc(base_address, row + i * 16, col + j * 16));
-      }
-    }
-  }
-
-  template <typename U>
-  __device__ inline void store_global(U* x, int N, int row, int col) {
-    MLX_UNROLL
-    for (int i = 0; i < TILES_Y; i++) {
-      MLX_UNROLL
-      for (int j = 0; j < TILES_X; j++) {
-        data[i * TILES_X + j].store_global(
-            x + (row + i * 16) * N + col + j * 16, N);
-      }
-    }
-  }
-};
-
 template <typename T, int ROWS_, int COLS_>
 struct SharedTile {
   static constexpr int ROWS = ROWS_;
@@ -282,8 +215,7 @@ struct SharedTile {
   static constexpr int TILES_Y = ROWS / 16;
   static constexpr int NUMEL = ROWS * COLS;
 
-  // Swizzle taken from ThunderKittens. Should be changed when we switch to
-  // cute Layouts.
+  // Swizzle taken from ThunderKittens.
   //
   // See inludes/types/shared/st.cuh
   //
@@ -296,10 +228,6 @@ struct SharedTile {
 
   T data[ROWS * COLS];
 
-  __device__ inline uint32_t base_addr() const {
-    return __cvta_generic_to_shared(&data[0]);
-  }
-
   // Return a pointer to the element at (row, col) using the swizzle.
   __device__ static inline T* ptr(T* ptr, int row, int col) {
     if constexpr (swizzle_bytes > 0) {
@@ -367,8 +295,6 @@ struct SharedTile {
 /**
  * Load the tile from global memory by loading 16 bytes at a time and storing
  * them immediately.
- *
- * Can also be used as a fallback for architectures before sm_80.
  */
 template <int NUM_WARPS, typename T, typename Tile>
 __device__ inline void load(Tile& tile, const T* x, int N) {
@@ -392,6 +318,33 @@ __device__ inline void load(Tile& tile, const T* x, int N) {
   }
 }
 
+/**
+ * Copy 16 bytes from the globale memory address pointed to by x to the smem
+ * address pointed to by row_address.
+ *
+ * A simple wrapper over the PTX.
+ */
+template <typename T>
+__device__ inline void cp_async_16(uint32_t row_address, const T* x) {
+  asm volatile(
+      "cp.async.ca.shared::cta.global [%0], [%1], 16;\n" ::"r"(row_address),
+      "l"(reinterpret_cast<const int4*>(x)));
+}
+
+/**
+ * Submit all the previous async copies to be executed.
+ */
+__device__ inline void cp_async_commit() {
+  asm volatile("cp.async.commit_group;\n" ::);
+}
+
+/**
+ * Wait for all the async copies to finish.
+ */
+__device__ inline void cp_async_wait_all() {
+  asm volatile("cp.async.wait_all;\n" ::);
+}
+
 /**
  * The asynchronous equivalent of load.
  *
@@ -425,17 +378,12 @@ load_async(Tile& tile, uint32_t base_address, const T* x, int N) {
 
   MLX_UNROLL
   for (int i = 0; i < NUM_LOADS_PER_THREAD; i++) {
-    cp_async<16>(
+    cp_async_16(
         tile.loc(base_address, row + i * STEP_ROWS, col * ELEMENTS_PER_LOAD),
         x + i * STEP_ROWS * N);
   }
 }
 
-/**
- * Same as load_async but checks if we can load the row.
- *
- * NOTE: It should be changed to use a predicated cp async instead.
- */
 template <int NUM_WARPS, typename T, typename Tile>
 __device__ inline void load_async_safe(
     Tile& tile,
@@ -458,7 +406,7 @@ __device__ inline void load_async_safe(
   MLX_UNROLL
   for (int i = 0; i < NUM_LOADS_PER_THREAD; i++) {
     if (row + i * STEP_ROWS < max_rows) {
-      cp_async<16>(
+      cp_async_16(
           tile.loc(base_address, row + i * STEP_ROWS, col * ELEMENTS_PER_LOAD),
           x + i * STEP_ROWS * N);
     } else {

mlx/backend/cuda/primitives.cpp

@@ -1,54 +0,0 @@
-// Copyright © 2025 Apple Inc.
-
-#include "mlx/distributed/primitives.h"
-#include "mlx/fast_primitives.h"
-#include "mlx/primitives.h"
-
-namespace mlx::core {
-
-#define NO_GPU_MULTI(func)                                             \
-  void func::eval_gpu(                                                 \
-      const std::vector<array>& inputs, std::vector<array>& outputs) { \
-    throw std::runtime_error(#func " has no CUDA implementation.");    \
-  }
-
-#define NO_GPU_USE_FALLBACK(func)     \
-  bool func::use_fallback(Stream s) { \
-    return true;                      \
-  }                                   \
-  NO_GPU_MULTI(func)
-
-#define NO_GPU(func)                                                  \
-  void func::eval_gpu(const std::vector<array>& inputs, array& out) { \
-    throw std::runtime_error(#func " has no CUDA implementation.");   \
-  }
-
-NO_GPU(BlockMaskedMM)
-NO_GPU(DynamicSlice)
-NO_GPU(DynamicSliceUpdate)
-NO_GPU(FFT)
-NO_GPU(GatherMM)
-NO_GPU(GatherQMM)
-NO_GPU(Hadamard)
-NO_GPU(Load)
-NO_GPU_MULTI(LUF)
-NO_GPU_MULTI(QRF)
-NO_GPU(QuantizedMatmul)
-NO_GPU(SegmentedMM)
-NO_GPU_MULTI(SVD)
-NO_GPU(Inverse)
-NO_GPU(Cholesky)
-NO_GPU_MULTI(Eig)
-NO_GPU_MULTI(Eigh)
-
-namespace fast {
-NO_GPU_MULTI(CustomKernel)
-} // namespace fast
-
-namespace distributed {
-NO_GPU_MULTI(AllGather)
-NO_GPU_MULTI(Send)
-NO_GPU_MULTI(Recv)
-} // namespace distributed
-
-} // namespace mlx::core

mlx/backend/cuda/primitives.cu

@@ -0,0 +1,103 @@
+// Copyright © 2025 Apple Inc.
+
+#include "mlx/backend/cuda/device.h"
+#include "mlx/backend/cuda/device/arange.cuh"
+#include "mlx/backend/cuda/device/fp16_math.cuh"
+#include "mlx/backend/cuda/kernel_utils.cuh"
+#include "mlx/distributed/primitives.h"
+#include "mlx/dtype_utils.h"
+#include "mlx/fast_primitives.h"
+#include "mlx/primitives.h"
+
+#include <nvtx3/nvtx3.hpp>
+#include <thrust/device_ptr.h>
+#include <thrust/transform.h>
+
+#include <cassert>
+
+namespace mlx::core {
+
+void Arange::eval_gpu(const std::vector<array>& inputs, array& out) {
+  nvtx3::scoped_range r("Arange::eval_gpu");
+  assert(inputs.size() == 0);
+  out.set_data(allocator::malloc(out.nbytes()));
+  if (out.size() == 0) {
+    return;
+  }
+  auto& encoder = cu::get_command_encoder(stream());
+  encoder.set_output_array(out);
+  auto capture = encoder.capture_context();
+  dispatch_int_float_types(out.dtype(), "Arange", [&](auto type_tag) {
+    using CTYPE = MLX_GET_TYPE(type_tag);
+    using OutType = cuda_type_t<CTYPE>;
+    CTYPE step =
+        static_cast<CTYPE>(start_ + step_) - static_cast<CTYPE>(start_);
+    thrust::transform(
+        cu::thrust_policy(encoder.stream()),
+        thrust::counting_iterator<uint32_t>(0),
+        thrust::counting_iterator<uint32_t>(out.data_size()),
+        thrust::device_pointer_cast(out.data<OutType>()),
+        cu::Arange<OutType>{
+            static_cast<OutType>(start_), static_cast<OutType>(step)});
+  });
+}
+
+bool fast::ScaledDotProductAttention::use_fallback(
+    const array& q,
+    const array& k,
+    const array& v,
+    bool has_mask,
+    bool has_arr_mask,
+    bool do_causal,
+    Stream s) {
+  return true;
+}
+
+#define NO_GPU_MULTI(func)                                             \
+  void func::eval_gpu(                                                 \
+      const std::vector<array>& inputs, std::vector<array>& outputs) { \
+    throw std::runtime_error(#func " has no CUDA implementation.");    \
+  }
+
+#define NO_GPU_USE_FALLBACK(func)     \
+  bool func::use_fallback(Stream s) { \
+    return true;                      \
+  }                                   \
+  NO_GPU_MULTI(func)
+
+#define NO_GPU(func)                                                  \
+  void func::eval_gpu(const std::vector<array>& inputs, array& out) { \
+    throw std::runtime_error(#func " has no CUDA implementation.");   \
+  }
+
+NO_GPU(BlockMaskedMM)
+NO_GPU(Convolution)
+NO_GPU(DynamicSlice)
+NO_GPU(DynamicSliceUpdate)
+NO_GPU(FFT)
+NO_GPU(GatherMM)
+NO_GPU(GatherQMM)
+NO_GPU(Hadamard)
+NO_GPU(Load)
+NO_GPU_MULTI(LUF)
+NO_GPU_MULTI(QRF)
+NO_GPU(SegmentedMM)
+NO_GPU_MULTI(SVD)
+NO_GPU(Inverse)
+NO_GPU(Cholesky)
+NO_GPU_MULTI(Eig)
+NO_GPU_MULTI(Eigh)
+
+namespace fast {
+NO_GPU(ScaledDotProductAttention)
+NO_GPU_MULTI(CustomKernel)
+} // namespace fast
+
+namespace distributed {
+NO_GPU_MULTI(AllReduce)
+NO_GPU_MULTI(AllGather)
+NO_GPU_MULTI(Send)
+NO_GPU_MULTI(Recv)
+} // namespace distributed
+
+} // namespace mlx::core

mlx/backend/cuda/quantized/affine_quantize.cu

@@ -256,7 +256,7 @@ void affine_quantize(
         using T = cuda_type_t<MLX_GET_TYPE(type_tag)>;
         auto kernel = cu::affine_quantize<T, group_size.value, bits.value>;
         auto [num_blocks, block_dims] =
-            get_launch_args(size, grid_shape, w.strides(), large);
+            get_launch_args(kernel, size, grid_shape, w.strides(), large);
         enc.add_kernel_node(
             kernel,
             num_blocks,
@@ -312,7 +312,7 @@ void affine_dequantize(
         using T = cuda_type_t<MLX_GET_TYPE(type_tag)>;
         auto kernel = cu::affine_dequantize<T, group_size.value, bits.value>;
         auto [num_blocks, block_dims] =
-            get_launch_args(size, grid_shape, w.strides(), large);
+            get_launch_args(kernel, size, grid_shape, w.strides(), large);
         enc.add_kernel_node(
             kernel,
             num_blocks,

mlx/backend/cuda/quantized/qmm.cu

@@ -0,0 +1,228 @@
+// Copyright © 2025 Apple Inc.
+
+#include "mlx/backend/cuda/device.h"
+#include "mlx/backend/cuda/kernel_utils.cuh"
+#include "mlx/backend/cuda/matmul/mma.cuh"
+#include "mlx/backend/cuda/matmul/tiles.cuh"
+#include "mlx/backend/cuda/quantized/quantized_utils.cuh"
+#include "mlx/dtype_utils.h"
+
+namespace mlx::core {
+
+namespace cu {
+
+template <int NUM_WARPS, int group_size, int bits, typename T, typename Tile>
+__device__ inline void load_quantized(
+    Tile& tile,
+    const uint8_t* x,
+    const T* scales,
+    const T* biases,
+    int N) {
+  constexpr int NUM_THREADS = NUM_WARPS * 32;
+  constexpr int ELEMENTS_PER_LOAD = sizeof(uint32_t) * get_pack_factor<bits>();
+  constexpr int NUM_LOADS = Tile::NUMEL / ELEMENTS_PER_LOAD;
+  constexpr int NUM_LOADS_PER_THREAD = NUM_LOADS / NUM_THREADS;
+  constexpr int NUM_LOADS_PER_ROW = Tile::COLS / ELEMENTS_PER_LOAD;
+  constexpr int STEP_ROWS = NUM_THREADS / NUM_LOADS_PER_ROW;
+  constexpr int MASK = (1 << bits) - 1;
+
+  const int row = threadIdx.x / NUM_LOADS_PER_ROW;
+  const int col = threadIdx.x % NUM_LOADS_PER_ROW;
+
+  const int Nx = N / get_pack_factor<bits>();
+  const int Ng = N / group_size;
+
+  x += row * Nx + col * (ELEMENTS_PER_LOAD / get_pack_factor<bits>());
+  scales += row * Ng + col * ELEMENTS_PER_LOAD / group_size;
+  biases += row * Ng + col * ELEMENTS_PER_LOAD / group_size;
+
+  MLX_UNROLL
+  for (int i = 0; i < NUM_LOADS_PER_THREAD; i++) {
+    T vs[ELEMENTS_PER_LOAD];
+    uint32_t w = *reinterpret_cast<const uint32_t*>(x + i * STEP_ROWS * Nx);
+    T s = scales[i * STEP_ROWS * Ng];
+    T b = biases[i * STEP_ROWS * Ng];
+    MLX_UNROLL
+    for (int j = 0; j < ELEMENTS_PER_LOAD; j++) {
+      vs[j] = static_cast<T>((w >> (j * bits)) & MASK) * s + b;
+    }
+    tile.store(vs, row + i * STEP_ROWS, col * ELEMENTS_PER_LOAD);
+  }
+}
+
+template <
+    typename T,
+    int BM,
+    int BN,
+    int BK,
+    int group_size,
+    int bits,
+    bool aligned_M>
+__global__ void qmm_t(
+    const T* x,
+    const uint8_t* w,
+    const T* scales,
+    const T* biases,
+    T* y,
+    int M,
+    int N,
+    int K) {
+  constexpr int WARPS_M = 2;
+  constexpr int WARPS_N = 4;
+  constexpr int NUM_WARPS = WARPS_M * WARPS_N;
+  constexpr int WARP_STEP_M = BM / WARPS_M;
+  constexpr int WARP_STEP_N = BN / WARPS_N;
+
+  const int warpid = threadIdx.x / 32;
+  const int laneid = threadIdx.x % 32;
+  const int wm = warpid / WARPS_N;
+  const int wn = warpid % WARPS_N;
+  const int offset_m = wm * WARP_STEP_M;
+  const int offset_n = wn * WARP_STEP_N;
+
+  extern __shared__ char shmem[];
+  SharedTile<T, BM, BK>(&xs)[1] = *(SharedTile<T, BM, BK>(*)[1])(&shmem[0]);
+  SharedTile<T, BN, BK>(&ws)[1] =
+      *(SharedTile<T, BN, BK>(*)[1])(&shmem[1 * sizeof(T) * BM * BK]);
+
+  RegisterTile<float, BM / WARPS_M, BN / WARPS_N> C;
+  RegisterTile<T, BM / WARPS_M, 16> A;
+  RegisterTile<T, BN / WARPS_N, 16> B;
+
+  const int max_rows = M - blockIdx.y * BM;
+
+  x += blockIdx.y * BM * K;
+  w += blockIdx.x * BN * K / get_pack_factor<bits>();
+  scales += blockIdx.x * BN * K / group_size;
+  biases += blockIdx.x * BN * K / group_size;
+  y += blockIdx.y * BM * N + blockIdx.x * BN;
+
+  C.fill(0);
+
+  int tic = 0;
+  uint32_t base_addr_xs[1], base_addr_ws[1];
+  base_addr_xs[0] = __cvta_generic_to_shared(&xs[0].data[0]);
+  base_addr_ws[0] = __cvta_generic_to_shared(&ws[0].data[0]);
+
+  if (aligned_M || max_rows >= BM) {
+    for (int k_block = 0; k_block < K; k_block += BK) {
+      load_async<NUM_WARPS>(xs[tic], base_addr_xs[tic], x + k_block, K);
+      cp_async_commit();
+      load_quantized<NUM_WARPS, group_size, bits>(
+          ws[tic],
+          w + k_block / get_pack_factor<bits>(),
+          scales + k_block / group_size,
+          biases + k_block / group_size,
+          K);
+      cp_async_wait_all();
+      __syncthreads();
+
+      MLX_UNROLL
+      for (int k = 0; k < BK / 16; k++) {
+        A.load(
+            xs[tic],
+            base_addr_xs[tic],
+            offset_m + laneid % 16,
+            k * 16 + laneid / 16 * 8);
+        B.load(
+            ws[tic],
+            base_addr_ws[tic],
+            offset_n + laneid % 16,
+            k * 16 + laneid / 16 * 8);
+        mma_t(C, A, B);
+      }
+    }
+    C.store_global(y, N, offset_m, offset_n);
+  } else {
+    for (int k_block = 0; k_block < K; k_block += BK) {
+      load_async_safe<NUM_WARPS>(
+          xs[tic], base_addr_xs[tic], x + k_block, K, max_rows);
+      cp_async_commit();
+      load_quantized<NUM_WARPS, group_size, bits>(
+          ws[tic],
+          w + k_block / get_pack_factor<bits>(),
+          scales + k_block / group_size,
+          biases + k_block / group_size,
+          K);
+      cp_async_wait_all();
+      __syncthreads();
+
+      MLX_UNROLL
+      for (int k = 0; k < BK / 16; k++) {
+        A.load(
+            xs[tic],
+            base_addr_xs[tic],
+            offset_m + laneid % 16,
+            k * 16 + laneid / 16 * 8);
+        B.load(
+            ws[tic],
+            base_addr_ws[tic],
+            offset_n + laneid % 16,
+            k * 16 + laneid / 16 * 8);
+        mma_t(C, A, B);
+      }
+    }
+    C.store_global_safe(y, N, offset_m, offset_n, max_rows);
+  }
+}
+
+} // namespace cu
+
+void qmm(
+    const array& x,
+    const array& w,
+    const array& scales,
+    const array& biases,
+    array& out,
+    bool transpose_,
+    int group_size_,
+    int bits_,
+    int M,
+    int N,
+    int K,
+    cu::CommandEncoder& enc,
+    const Stream& s) {
+  if (x.dtype() != bfloat16) {
+    throw std::invalid_argument("[qmm] Only bfloat16 is supported for now");
+  }
+  if (!transpose_) {
+    throw std::invalid_argument(
+        "[qmm] Only transposed matmul is supported for now");
+  }
+
+  dispatch_float_types(x.dtype(), "qmm", [&](auto type_tag) {
+    dispatch_groups(group_size_, [&](auto group_size) {
+      dispatch_bits(bits_, [&](auto bits) {
+        using DataType = cuda_type_t<MLX_GET_TYPE(type_tag)>;
+
+        constexpr int BM = 128;
+        constexpr int BN = 128;
+        constexpr int BK = 32;
+        auto kernel =
+            cu::qmm_t<DataType, BM, BN, BK, group_size.value, bits.value, true>;
+        if (M % BM != 0) {
+          kernel = cu::
+              qmm_t<DataType, BM, BN, BK, group_size.value, bits.value, false>;
+        }
+
+        dim3 grid((N + BN - 1) / BN, (M + BM - 1) / BM);
+
+        enc.add_kernel_node(
+            kernel,
+            grid,
+            2 * 4 * 32,
+            1 * sizeof(DataType) * (BM * BK + BN * BK),
+            x.data<DataType>(),
+            w.data<uint8_t>(),
+            scales.data<DataType>(),
+            biases.data<DataType>(),
+            out.data<DataType>(),
+            M,
+            N,
+            K);
+      });
+    });
+  });
+}
+
+} // namespace mlx::core

mlx/backend/cuda/quantized/quantized.cu

@@ -1,10 +1,14 @@
 // Copyright © 2025 Apple Inc.
 
-#include "mlx/backend/cuda/quantized/quantized.h"
 #include "mlx/backend/cuda/device.h"
+#include "mlx/backend/cuda/kernel_utils.cuh"
+#include "mlx/backend/cuda/quantized/quantized.cuh"
 #include "mlx/backend/gpu/copy.h"
+#include "mlx/dtype_utils.h"
 #include "mlx/fast_primitives.h"
 
+#include <cooperative_groups.h>
+#include <cooperative_groups/reduce.h>
 #include <nvtx3/nvtx3.hpp>
 
 namespace mlx::core {
@@ -28,28 +32,57 @@ inline array ensure_row_contiguous_matrix(
     const array& x,
     cu::CommandEncoder& enc,
     const Stream& s) {
-  if (x.ndim() < 2) {
-    if (x.strides()[0] == 1) {
-      return x;
-    }
+  auto stride_0 = x.strides()[x.ndim() - 2];
+  auto stride_1 = x.strides()[x.ndim() - 1];
+  if (stride_0 == x.shape(-1) && stride_1 == 1) {
+    return x;
   } else {
-    auto stride_0 = x.strides()[x.ndim() - 2];
-    auto stride_1 = x.strides()[x.ndim() - 1];
-    if (stride_0 == x.shape(-1) && stride_1 == 1) {
-      return x;
-    }
+    array x_copy = contiguous_copy_gpu(x, s);
+    enc.add_temporary(x_copy);
+    return x_copy;
   }
-  array x_copy = contiguous_copy_gpu(x, s);
-  enc.add_temporary(x_copy);
-  return x_copy;
 }
 
 } // namespace
 
+void QuantizedMatmul::eval_gpu(const std::vector<array>& inputs, array& out) {
+  auto& s = stream();
+  auto& d = cu::device(s.device);
+  auto& enc = d.get_command_encoder(s);
+
+  out.set_data(allocator::malloc(out.nbytes()));
+
+  // Make sure the last two dims of x and w, s, b are contiguous. This should
+  // be relaxed for x.
+  array x = ensure_row_contiguous_matrix(inputs[0], enc, s);
+  array w = ensure_row_contiguous_matrix(inputs[1], enc, s);
+  array scales = ensure_row_contiguous_matrix(inputs[2], enc, s);
+  array biases = ensure_row_contiguous_matrix(inputs[3], enc, s);
+
+  // Extract the matmul shapes
+  bool non_batched = w.ndim() == 2 && x.flags().row_contiguous;
+  int K = x.shape(-1);
+  int M = non_batched ? x.size() / K : x.shape(-2);
+  int N = out.shape(-1);
+
+  qmm(x,
+      w,
+      scales,
+      biases,
+      out,
+      transpose_,
+      group_size_,
+      bits_,
+      M,
+      N,
+      K,
+      enc,
+      s);
+}
+
 void fast::AffineQuantize::eval_gpu(
     const std::vector<array>& inputs,
     std::vector<array>& outputs) {
-  nvtx3::scoped_range r("AffineQuantize::eval_gpu");
   auto& s = stream();
   auto& d = cu::device(s.device);
   auto& enc = d.get_command_encoder(s);

mlx/backend/cuda/quantized/quantized.cuh

@@ -24,4 +24,19 @@ void affine_dequantize(
     cu::CommandEncoder& enc,
     const Stream& s);
 
+void qmm(
+    const array& x,
+    const array& w,
+    const array& scales,
+    const array& biases,
+    array& out,
+    bool transpose_,
+    int group_size_,
+    int bits_,
+    int M,
+    int N,
+    int K,
+    cu::CommandEncoder& enc,
+    const Stream& s);
+
 } // namespace mlx::core

mlx/backend/cuda/reduce.cu

@@ -5,6 +5,8 @@
 #include "mlx/backend/gpu/copy.h"
 
 #include <nvtx3/nvtx3.hpp>
+#include <thrust/device_ptr.h>
+#include <thrust/fill.h>
 
 #include <cassert>
 

mlx/backend/cuda/rms_norm.cu

@@ -1,6 +1,7 @@
 // Copyright © 2025 Apple Inc.
 
 #include "mlx/backend/cuda/device.h"
+#include "mlx/backend/cuda/iterators/strided_iterator.cuh"
 #include "mlx/backend/cuda/kernel_utils.cuh"
 #include "mlx/backend/cuda/reduce/reduce.cuh"
 #include "mlx/backend/gpu/copy.h"
@@ -10,6 +11,8 @@
 #include <cooperative_groups.h>
 #include <cooperative_groups/reduce.h>
 #include <nvtx3/nvtx3.hpp>
+#include <cub/block/block_load.cuh>
+#include <cub/block/block_reduce.cuh>
 
 namespace mlx::core {
 
@@ -55,7 +58,7 @@ __global__ void rms_norm(
     const T* w,
     T* out,
     float eps,
-    uint32_t axis_size,
+    int32_t axis_size,
     int64_t w_stride) {
   auto grid = cg::this_grid();
   auto block = cg::this_thread_block();
@@ -70,8 +73,8 @@ __global__ void rms_norm(
   float normalizer = 0;
   for (int r = 0; r < cuda::ceil_div(axis_size, BLOCK_DIM * N_READS); ++r) {
     auto index = r * BLOCK_DIM + block.thread_rank();
-    auto xn = load_vector<N_READS>(x, index, axis_size, T(0));
-#pragma unroll
+    T xn[N_READS];
+    cub::LoadDirectBlocked(index, x, xn, axis_size, cast_to<T>(0));
     for (int i = 0; i < N_READS; ++i) {
       float t = static_cast<float>(xn[i]);
       normalizer += t * t;
@@ -83,14 +86,15 @@ __global__ void rms_norm(
   // Outputs.
   for (int r = 0; r < cuda::ceil_div(axis_size, BLOCK_DIM * N_READS); ++r) {
     auto index = r * BLOCK_DIM + block.thread_rank();
-    auto xn = load_vector<N_READS>(x, index, axis_size, T(0));
-    auto wn = load_vector<N_READS>(w, index, axis_size, w_stride, T(0));
-#pragma unroll
+    T xn[N_READS];
+    T wn[N_READS];
+    cub::LoadDirectBlocked(index, x, xn, axis_size);
+    cub::LoadDirectBlocked(index, strided_iterator(w, w_stride), wn, axis_size);
     for (int i = 0; i < N_READS; ++i) {
-      float y = static_cast<float>(xn[i]) * normalizer;
-      xn[i] = wn[i] * static_cast<T>(y);
+      float norm = static_cast<float>(xn[i]) * normalizer;
+      xn[i] = wn[i] * static_cast<T>(norm);
     }
-    store_vector<N_READS>(out, index, xn, axis_size);
+    cub::StoreDirectBlocked(index, out, xn, axis_size);
   }
 }
 
@@ -122,10 +126,13 @@ __global__ void rms_norm_vjp(
   // Normalizer.
   float2 factors = {};
   for (int r = 0; r < cuda::ceil_div(axis_size, BLOCK_DIM * N_READS); ++r) {
+    T xn[N_READS];
+    T wn[N_READS] = {};
+    T gn[N_READS] = {};
     auto index = r * BLOCK_DIM + block.thread_rank();
-    auto xn = load_vector<N_READS>(x, index, axis_size, T(0));
-    auto gn = load_vector<N_READS>(g, index, axis_size, T(0));
-    auto wn = load_vector<N_READS>(w, index, axis_size, w_stride, T(0));
+    cub::LoadDirectBlocked(index, x, xn, axis_size, cast_to<T>(0));
+    cub::LoadDirectBlocked(index, g, gn, axis_size);
+    cub::LoadDirectBlocked(index, strided_iterator(w, w_stride), wn, axis_size);
     for (int i = 0; i < N_READS; i++) {
       float t = static_cast<float>(xn[i]);
       float wi = wn[i];
@@ -142,9 +149,12 @@ __global__ void rms_norm_vjp(
   // Outputs.
   for (int r = 0; r < cuda::ceil_div(axis_size, BLOCK_DIM * N_READS); ++r) {
     auto index = r * BLOCK_DIM + block.thread_rank();
-    auto xn = load_vector<N_READS>(x, index, axis_size, T(0));
-    auto gn = load_vector<N_READS>(g, index, axis_size, T(0));
-    auto wn = load_vector<N_READS>(w, index, axis_size, w_stride, T(0));
+    T xn[N_READS];
+    T wn[N_READS];
+    T gn[N_READS];
+    cub::LoadDirectBlocked(index, x, xn, axis_size);
+    cub::LoadDirectBlocked(index, g, gn, axis_size);
+    cub::LoadDirectBlocked(index, strided_iterator(w, w_stride), wn, axis_size);
     for (int i = 0; i < N_READS; i++) {
       float xi = xn[i];
       float wi = wn[i];
@@ -154,9 +164,9 @@ __global__ void rms_norm_vjp(
         wn[i] = static_cast<T>(gi * xi * normalizer);
       }
     }
-    store_vector<N_READS>(gx, index, xn, axis_size);
+    cub::StoreDirectBlocked(index, gx, xn, axis_size);
     if constexpr (HAS_W) {
-      store_vector<N_READS>(gw, index, wn, axis_size);
+      cub::StoreDirectBlocked(index, gw, wn, axis_size);
     }
   }
 }
@@ -214,9 +224,9 @@ void RMSNorm::eval_gpu(
   encoder.set_input_array(w);
   encoder.set_output_array(out);
   dispatch_float_types(out.dtype(), "rms_norm", [&](auto type_tag) {
-    using DataType = cuda_type_t<MLX_GET_TYPE(type_tag)>;
-    constexpr int N_READS = 16 / sizeof(DataType);
+    constexpr uint32_t N_READS = 4;
     dispatch_block_dim(cuda::ceil_div(axis_size, N_READS), [&](auto block_dim) {
+      using DataType = cuda_type_t<MLX_GET_TYPE(type_tag)>;
       auto kernel = cu::rms_norm<DataType, block_dim(), N_READS>;
       encoder.add_kernel_node(
           kernel,
@@ -304,10 +314,11 @@ void RMSNormVJP::eval_gpu(
   encoder.set_output_array(gw_temp);
   dispatch_float_types(gx.dtype(), "rms_norm_vjp", [&](auto type_tag) {
     dispatch_bool(has_w, [&](auto has_w_constant) {
-      using DataType = cuda_type_t<MLX_GET_TYPE(type_tag)>;
-      constexpr int N_READS = 16 / sizeof(DataType);
+      constexpr int N_READS = 4;
       dispatch_block_dim(
           cuda::ceil_div(axis_size, N_READS), [&](auto block_dim) {
+            using DataType = cuda_type_t<MLX_GET_TYPE(type_tag)>;
+            constexpr int N_READS = 4;
             auto kernel = cu::rms_norm_vjp<
                 DataType,
                 has_w_constant.value,

mlx/backend/cuda/scaled_dot_product_attention.cu

@@ -1,781 +0,0 @@
-// Copyright © 2025 Apple Inc.
-
-#include "mlx/backend/cuda/device.h"
-#include "mlx/backend/cuda/device/config.h"
-#include "mlx/backend/cuda/device/utils.cuh"
-#include "mlx/backend/cuda/kernel_utils.cuh"
-#include "mlx/backend/cuda/lru_cache.h"
-#include "mlx/backend/gpu/copy.h"
-#include "mlx/dtype_utils.h"
-#include "mlx/fast_primitives.h"
-#include "mlx/transforms_impl.h"
-
-#include <nvtx3/nvtx3.hpp>
-
-#include <cooperative_groups.h>
-#include <cooperative_groups/reduce.h>
-
-namespace mlx::core {
-
-namespace cu {
-
-namespace cg = cooperative_groups;
-
-#define PRAGMA_LOOP_UNROLL #pragma unroll
-
-struct AttnParams {
-  int B;
-  int H;
-  int D;
-
-  int qL;
-  int kL;
-
-  int gqa_factor;
-  float scale;
-
-  int64_t Q_strides[3];
-  int64_t K_strides[3];
-  int64_t V_strides[3];
-  int64_t O_strides[3];
-};
-
-template <typename T, bool do_causal, int D>
-__global__ void kernel_sdpav_1pass(
-    const T* Q,
-    const T* K,
-    const T* V,
-    T* O,
-    __grid_constant__ const AttnParams params) {
-  constexpr int BN = 32;
-  constexpr int BD = 32;
-
-  constexpr int v_per_thread = D / BD;
-
-  const int inner_k_stride = BN * int(params.K_strides[2]);
-  const int inner_v_stride = BN * int(params.V_strides[2]);
-
-  typedef float U;
-
-  U q[v_per_thread];
-  U k[v_per_thread];
-  U o[v_per_thread];
-
-  __shared__ U outputs[BN][BD + 1];
-  __shared__ U max_scores[BN];
-  __shared__ U sum_exp_scores[BN];
-
-  const U scale_log2 = params.scale * 1.44269504089f;
-
-  auto block = cg::this_thread_block();
-  auto warp = cg::tiled_partition<32>(block);
-
-  const int lane_idx = warp.thread_rank();
-  const int warp_idx = warp.meta_group_rank();
-
-  // Adjust to thread block and thread
-  const int batch_idx = blockIdx.z;
-  const int head_idx = blockIdx.x;
-  const int kv_head_idx = head_idx / params.gqa_factor;
-
-  const int q_seq_idx = blockIdx.y;
-  const int kv_seq_idx = warp_idx;
-
-  Q += batch_idx * params.Q_strides[0] + // Batch
-      head_idx * params.Q_strides[1] + // Head
-      q_seq_idx * params.Q_strides[2]; // Sequence
-
-  K += batch_idx * params.K_strides[0] + // Batch
-      kv_head_idx * params.K_strides[1] + // Head
-      kv_seq_idx * params.K_strides[2]; // Sequence
-
-  V += batch_idx * params.V_strides[0] + // Batch
-      kv_head_idx * params.V_strides[1] + // Head
-      kv_seq_idx * params.V_strides[2]; // Sequence
-
-  O += batch_idx * params.O_strides[0] + // Batch
-      head_idx * params.O_strides[1] + // Head
-      q_seq_idx * params.O_strides[2]; // Sequence
-
-  // Read the query and 0 the output accumulator
-  PRAGMA_LOOP_UNROLL
-  for (int i = 0; i < v_per_thread; i++) {
-    q[i] = scale_log2 * static_cast<U>(Q[v_per_thread * lane_idx + i]);
-  }
-
-  PRAGMA_LOOP_UNROLL
-  for (int i = 0; i < v_per_thread; i++) {
-    o[i] = 0.f;
-  }
-
-  U max_score = -INFINITY;
-  U sum_exp_score = 0.f;
-
-  // For each key
-  for (int i = kv_seq_idx; i < params.kL; i += BN) {
-    bool use_key = true;
-    if constexpr (do_causal) {
-      use_key = i <= (params.kL - params.qL + q_seq_idx);
-    }
-
-    if (use_key) {
-      // Read the key
-      PRAGMA_LOOP_UNROLL
-      for (int j = 0; j < v_per_thread; j++) {
-        k[j] = K[v_per_thread * lane_idx + j];
-      }
-
-      // Compute the i-th score
-      U score = 0.f;
-      PRAGMA_LOOP_UNROLL
-      for (int j = 0; j < v_per_thread; j++) {
-        score += q[j] * k[j];
-      }
-
-      // Warp sum
-      score = cg::reduce(warp, score, cg::plus<U>());
-
-      // Update the accumulators
-      U new_max = max(max_score, score);
-      U factor = exp2f(max_score - new_max);
-      U exp_score = exp2f(score - new_max);
-
-      max_score = new_max;
-      sum_exp_score = sum_exp_score * factor + exp_score;
-
-      // Update the output accumulator
-      PRAGMA_LOOP_UNROLL
-      for (int j = 0; j < v_per_thread; j++) {
-        o[j] = o[j] * factor +
-            exp_score * static_cast<U>(V[v_per_thread * lane_idx + j]);
-      }
-    }
-
-    // Move the pointers to the next kv
-    K += inner_k_stride;
-    V += inner_v_stride;
-  }
-
-  if (lane_idx == 0) {
-    max_scores[warp_idx] = max_score;
-    sum_exp_scores[warp_idx] = sum_exp_score;
-  }
-  block.sync();
-
-  max_score = max_scores[lane_idx];
-  U new_max = cg::reduce(warp, max_score, cg::greater<U>());
-  U factor = exp2f(max_score - new_max);
-  sum_exp_score =
-      cg::reduce(warp, sum_exp_scores[lane_idx] * factor, cg::plus<U>());
-  sum_exp_score = __frcp_rn(sum_exp_score);
-
-  // Now we need to aggregate all the outputs
-  PRAGMA_LOOP_UNROLL
-  for (int i = 0; i < v_per_thread; i++) {
-    outputs[lane_idx][warp_idx] = o[i];
-    block.sync();
-    U ot = outputs[warp_idx][lane_idx] * factor;
-    o[i] = cg::reduce(warp, ot, cg::plus<U>()) * sum_exp_score;
-    block.sync();
-  }
-
-  // And write the output
-  if (lane_idx == 0) {
-    PRAGMA_LOOP_UNROLL
-    for (int i = 0; i < v_per_thread; i++) {
-      O[v_per_thread * warp_idx + i] = static_cast<T>(o[i]);
-    }
-  }
-}
-
-template <typename T, bool do_causal, int D>
-__global__ void kernel_sdpav_2pass_1(
-    const T* Q,
-    const T* K,
-    const T* V,
-    float* partials,
-    float* sums,
-    float* maxs,
-    __grid_constant__ const AttnParams params) {
-  constexpr int BN = 8;
-  constexpr int BD = 32;
-  constexpr int blocks = 32;
-
-  constexpr int v_per_thread = D / BD;
-
-  const int inner_k_stride = blocks * BN * int(params.K_strides[2]);
-  const int inner_v_stride = blocks * BN * int(params.V_strides[2]);
-
-  typedef float U;
-
-  U q[v_per_thread];
-  U k[v_per_thread];
-  U o[v_per_thread];
-
-  __shared__ U outputs[BN][BD + 1];
-  __shared__ U max_scores[BN];
-  __shared__ U sum_exp_scores[BN];
-
-  const U scale_log2 = params.scale * 1.44269504089f;
-
-  auto block = cg::this_thread_block();
-  auto warp = cg::tiled_partition<32>(block);
-
-  const int lane_idx = warp.thread_rank();
-  const int warp_idx = warp.meta_group_rank();
-
-  // Adjust to thread block and thread
-  const int batch_idx = blockIdx.z / blocks;
-  const int block_idx = blockIdx.z % blocks;
-  const int head_idx = blockIdx.x;
-  const int kv_head_idx = head_idx / params.gqa_factor;
-
-  const int q_seq_idx = blockIdx.y;
-  const int kv_seq_idx = block_idx * BN + warp_idx;
-
-  Q += batch_idx * params.Q_strides[0] + // Batch
-      head_idx * params.Q_strides[1] + // Head
-      q_seq_idx * params.Q_strides[2]; // Sequence
-
-  K += batch_idx * params.K_strides[0] + // Batch
-      kv_head_idx * params.K_strides[1] + // Head
-      kv_seq_idx * params.K_strides[2]; // Sequence
-
-  V += batch_idx * params.V_strides[0] + // Batch
-      kv_head_idx * params.V_strides[1] + // Head
-      kv_seq_idx * params.V_strides[2]; // Sequence
-
-  const int p_stride_s = blocks;
-  const int p_stride_h = params.qL * p_stride_s;
-  const int p_stride_b = params.H * p_stride_h;
-  const int p_offset = batch_idx * p_stride_b + // Batch
-      head_idx * p_stride_h + // Head
-      q_seq_idx * p_stride_s + // Sequence
-      block_idx; // Block
-
-  partials += p_offset * D;
-  sums += p_offset;
-  maxs += p_offset;
-
-  // Read the query and 0 the output accumulator
-  PRAGMA_LOOP_UNROLL
-  for (int i = 0; i < v_per_thread; i++) {
-    q[i] = scale_log2 * static_cast<U>(Q[v_per_thread * lane_idx + i]);
-  }
-
-  PRAGMA_LOOP_UNROLL
-  for (int i = 0; i < v_per_thread; i++) {
-    o[i] = 0.f;
-  }
-
-  U max_score = -1e9;
-  U sum_exp_score = 0.f;
-
-  // For each key
-  for (int i = kv_seq_idx; i < params.kL; i += blocks * BN) {
-    bool use_key = true;
-    if constexpr (do_causal) {
-      use_key = i <= (params.kL - params.qL + q_seq_idx);
-    }
-
-    if (use_key) {
-      // Read the key
-      PRAGMA_LOOP_UNROLL
-      for (int j = 0; j < v_per_thread; j++) {
-        k[j] = K[v_per_thread * lane_idx + j];
-      }
-
-      // Compute the i-th score
-      U score = 0.f;
-      PRAGMA_LOOP_UNROLL
-      for (int j = 0; j < v_per_thread; j++) {
-        score += q[j] * k[j];
-      }
-
-      // Warp sum
-      score = cg::reduce(warp, score, cg::plus<U>());
-
-      // Update the accumulators
-      U new_max = max(max_score, score);
-      U factor = exp2f(max_score - new_max);
-      U exp_score = exp2f(score - new_max);
-
-      max_score = new_max;
-      sum_exp_score = sum_exp_score * factor + exp_score;
-
-      // Update the output accumulator
-      PRAGMA_LOOP_UNROLL
-      for (int j = 0; j < v_per_thread; j++) {
-        o[j] = o[j] * factor +
-            exp_score * static_cast<U>(V[v_per_thread * lane_idx + j]);
-      }
-    }
-
-    // Move the pointers to the next kv
-    K += inner_k_stride;
-    V += inner_v_stride;
-  }
-
-  if (lane_idx == 0) {
-    max_scores[warp_idx] = max_score;
-    sum_exp_scores[warp_idx] = sum_exp_score;
-  }
-
-  block.sync();
-
-  max_score = (lane_idx < BN) ? max_scores[lane_idx] : -1e9;
-  U new_max = cg::reduce(warp, max_score, cg::greater<U>());
-  U factor = exp2f(max_score - new_max);
-  sum_exp_score = (lane_idx < BN) ? sum_exp_scores[lane_idx] : 0.f;
-  sum_exp_score = cg::reduce(warp, sum_exp_score * factor, cg::plus<U>());
-
-  // Write the sum and new max
-  if (warp_idx == 0) {
-    sums[0] = sum_exp_score;
-    maxs[0] = new_max;
-  }
-
-  // Now we need to aggregate all the outputs
-  auto ff = exp2f(max_scores[warp_idx] - new_max);
-  PRAGMA_LOOP_UNROLL
-  for (int i = 0; i < v_per_thread; i++) {
-    outputs[warp_idx][lane_idx] = o[i] * ff;
-    block.sync();
-
-    if (warp_idx == 0) {
-      U ot = outputs[0][lane_idx];
-      PRAGMA_LOOP_UNROLL
-      for (int j = 1; j < BN; j++) {
-        ot += outputs[j][lane_idx];
-        warp.sync();
-      }
-      o[i] = ot;
-    }
-    block.sync();
-  }
-
-  if (warp_idx == 0) {
-    PRAGMA_LOOP_UNROLL
-    for (int i = 0; i < v_per_thread; i++) {
-      partials[v_per_thread * lane_idx + i] = o[i];
-    }
-  }
-}
-
-template <typename T, bool do_causal, int D>
-__global__ void kernel_sdpav_2pass_2(
-    const float* partials,
-    const float* sums,
-    const float* maxs,
-    T* O,
-    __grid_constant__ const AttnParams params) {
-  constexpr int BN = 32;
-  constexpr int BD = 32;
-  constexpr int blocks = 32;
-
-  constexpr int v_per_thread = D / BD;
-
-  typedef float U;
-
-  U o[v_per_thread];
-  __shared__ U outputs[BN][BD + 1];
-
-  auto block = cg::this_thread_block();
-  auto warp = cg::tiled_partition<32>(block);
-
-  const int lane_idx = warp.thread_rank();
-  const int warp_idx = warp.meta_group_rank();
-
-  // Adjust to thread block and thread
-  const int batch_idx = blockIdx.z;
-  const int head_idx = blockIdx.x;
-  const int q_seq_idx = blockIdx.y;
-
-  const int p_stride_s = blocks;
-  const int p_stride_h = params.qL * p_stride_s;
-  const int p_stride_b = params.H * p_stride_h;
-  const int p_offset = batch_idx * p_stride_b + // Batch
-      head_idx * p_stride_h + // Head
-      q_seq_idx * p_stride_s; // Sequence
-
-  partials += p_offset * D + warp_idx * D;
-  sums += p_offset;
-  maxs += p_offset;
-
-  O += batch_idx * params.O_strides[0] + // Batch
-      head_idx * params.O_strides[1] + // Head
-      q_seq_idx * params.O_strides[2]; // Sequence
-
-  U max_score = maxs[lane_idx];
-  U new_max = cg::reduce(warp, max_score, cg::greater<U>());
-  U factor = exp2f(max_score - new_max);
-  U sum_exp_score = cg::reduce(warp, sums[lane_idx] * factor, cg::plus<U>());
-  sum_exp_score = __frcp_rn(sum_exp_score);
-
-  PRAGMA_LOOP_UNROLL
-  for (int i = 0; i < v_per_thread; i++) {
-    o[i] = partials[v_per_thread * lane_idx + i];
-  }
-
-  // Now we need to aggregate all the outputs
-  PRAGMA_LOOP_UNROLL
-  for (int i = 0; i < v_per_thread; i++) {
-    outputs[lane_idx][warp_idx] = o[i];
-    block.sync();
-    U ot = outputs[warp_idx][lane_idx] * factor;
-    o[i] = cg::reduce(warp, ot, cg::plus<U>()) * sum_exp_score;
-    block.sync();
-  }
-
-  // And write the output
-  if (lane_idx == 0) {
-    PRAGMA_LOOP_UNROLL
-    for (int i = 0; i < v_per_thread; i++) {
-      O[v_per_thread * warp_idx + i] = static_cast<T>(o[i]);
-    }
-  }
-}
-
-} // namespace cu
-
-namespace {
-
-template <typename F>
-void dispatch_headdim(int n, F&& f) {
-  switch (n) {
-    case 64:
-      f(std::integral_constant<int, 64>{});
-      break;
-    case 96:
-      f(std::integral_constant<int, 96>{});
-      break;
-    case 128:
-      f(std::integral_constant<int, 128>{});
-      break;
-  }
-}
-
-void sdpa_vector_1pass_fallback(
-    const Stream& s,
-    cu::CommandEncoder& encoder,
-    const array& q,
-    const array& k,
-    const array& v,
-    const float scale,
-    array& o,
-    bool do_causal_ = false) {
-  encoder.set_input_array(q);
-  encoder.set_input_array(k);
-  encoder.set_input_array(v);
-  encoder.set_output_array(o);
-
-  cu::AttnParams params{
-      /* int B = */ q.shape(0),
-      /* int H = */ q.shape(1),
-      /* int D = */ q.shape(3),
-
-      /* int qL = */ q.shape(2),
-      /* int kL = */ k.shape(2),
-
-      /* int gqa_factor = */ q.shape(1) / k.shape(1),
-      /* float scale = */ scale,
-
-      /* int64_t Q_strides[3] = */ {q.strides(0), q.strides(1), q.strides(2)},
-      /* int64_t K_strides[3] = */ {k.strides(0), k.strides(1), k.strides(2)},
-      /* int64_t V_strides[3] = */ {v.strides(0), v.strides(1), v.strides(2)},
-      /* int64_t O_strides[3] = */ {o.strides(0), o.strides(1), o.strides(2)}};
-
-  dim3 grid_dim(params.H, params.qL, params.B);
-  dim3 block_dim(1024, 1, 1);
-
-  dispatch_float_types(o.dtype(), "kernel_sdpav_1pass", [&](auto type_tag) {
-    dispatch_bool(do_causal_, [&](auto do_causal) {
-      dispatch_headdim(params.D, [&](auto headdim) {
-        using DataType = cuda_type_t<MLX_GET_TYPE(type_tag)>;
-
-        auto kernel =
-            cu::kernel_sdpav_1pass<DataType, do_causal.value, headdim.value>;
-        encoder.add_kernel_node(
-            kernel,
-            grid_dim,
-            block_dim,
-            0,
-            q.data<DataType>(),
-            k.data<DataType>(),
-            v.data<DataType>(),
-            o.data<DataType>(),
-            params);
-      });
-    });
-  });
-}
-
-void sdpa_vector_2pass_fallback(
-    const Stream& s,
-    cu::CommandEncoder& encoder,
-    const array& q,
-    const array& k,
-    const array& v,
-    const float scale,
-    array& o,
-    bool do_causal_ = false) {
-  cu::AttnParams params{
-      /* int B = */ q.shape(0),
-      /* int H = */ q.shape(1),
-      /* int D = */ q.shape(3),
-
-      /* int qL = */ q.shape(2),
-      /* int kL = */ k.shape(2),
-
-      /* int gqa_factor = */ q.shape(1) / k.shape(1),
-      /* float scale = */ scale,
-
-      /* int64_t Q_strides[3] = */ {q.strides(0), q.strides(1), q.strides(2)},
-      /* int64_t K_strides[3] = */ {k.strides(0), k.strides(1), k.strides(2)},
-      /* int64_t V_strides[3] = */ {v.strides(0), v.strides(1), v.strides(2)},
-      /* int64_t O_strides[3] = */ {o.strides(0), o.strides(1), o.strides(2)}};
-
-  // Allocate the intermediates
-  int blocks = 32;
-
-  Shape intermediate_shape;
-  intermediate_shape.reserve(o.ndim() + 1);
-  intermediate_shape.insert(
-      intermediate_shape.end(), o.shape().begin(), o.shape().end() - 1);
-  intermediate_shape.push_back(blocks);
-  intermediate_shape.push_back(o.shape().back());
-
-  array intermediate(intermediate_shape, float32, nullptr, {});
-  intermediate_shape.pop_back();
-  array sums(intermediate_shape, float32, nullptr, {});
-  array maxs(std::move(intermediate_shape), float32, nullptr, {});
-
-  intermediate.set_data(allocator::malloc(intermediate.nbytes()));
-  sums.set_data(allocator::malloc(sums.nbytes()));
-  maxs.set_data(allocator::malloc(maxs.nbytes()));
-
-  encoder.add_temporary(intermediate);
-  encoder.add_temporary(sums);
-  encoder.add_temporary(maxs);
-
-  dispatch_float_types(o.dtype(), "kernel_sdpav_2pass", [&](auto type_tag) {
-    dispatch_bool(do_causal_, [&](auto do_causal) {
-      dispatch_headdim(params.D, [&](auto headdim) {
-        using DataType = cuda_type_t<MLX_GET_TYPE(type_tag)>;
-
-        {
-          auto kernel = cu::
-              kernel_sdpav_2pass_1<DataType, do_causal.value, headdim.value>;
-
-          encoder.set_input_array(q);
-          encoder.set_input_array(k);
-          encoder.set_input_array(v);
-          encoder.set_output_array(intermediate);
-          encoder.set_output_array(sums);
-          encoder.set_output_array(maxs);
-
-          dim3 grid_dim(params.H, params.qL, params.B * 32);
-          dim3 block_dim(8 * 32, 1, 1);
-
-          encoder.add_kernel_node(
-              kernel,
-              grid_dim,
-              block_dim,
-              0,
-              q.data<DataType>(),
-              k.data<DataType>(),
-              v.data<DataType>(),
-              intermediate.data<float>(),
-              sums.data<float>(),
-              maxs.data<float>(),
-              params);
-        }
-
-        {
-          auto kernel = cu::
-              kernel_sdpav_2pass_2<DataType, do_causal.value, headdim.value>;
-
-          encoder.set_input_array(intermediate);
-          encoder.set_input_array(sums);
-          encoder.set_input_array(maxs);
-          encoder.set_output_array(o);
-
-          dim3 grid_dim(params.H, params.qL, params.B);
-          dim3 block_dim(1024, 1, 1);
-
-          encoder.add_kernel_node(
-              kernel,
-              grid_dim,
-              block_dim,
-              0,
-              intermediate.data<float>(),
-              sums.data<float>(),
-              maxs.data<float>(),
-              o.data<DataType>(),
-              params);
-        }
-      });
-    });
-  });
-}
-
-void sdpa_vector_fallback(
-    const Stream& s,
-    cu::CommandEncoder& encoder,
-    const array& q,
-    const array& k,
-    const array& v,
-    const float scale,
-    array& o,
-    bool do_causal_ = false) {
-  int kL = k.shape(2);
-
-  if (kL > 1024) {
-    return sdpa_vector_2pass_fallback(
-        s, encoder, q, k, v, scale, o, do_causal_);
-  } else {
-    return sdpa_vector_1pass_fallback(
-        s, encoder, q, k, v, scale, o, do_causal_);
-  }
-}
-
-} // namespace
-
-namespace fast {
-
-bool ScaledDotProductAttention::use_fallback(
-    const array& q,
-    const array& k,
-    const array& v,
-    bool has_mask,
-    bool has_arr_mask,
-    bool do_causal,
-    Stream s) {
-  if (detail::in_grad_tracing()) {
-    return true;
-  }
-  if (s.device == Device::cpu) {
-    return true;
-  }
-
-  const int value_head_dim = v.shape(-1);
-  const int query_head_dim = q.shape(-1);
-  const int query_sequence_length = q.shape(2);
-  const int key_sequence_length = k.shape(2);
-
-  const bool sdpa_supported_head_dim = query_head_dim == value_head_dim &&
-      (query_head_dim == 64 || query_head_dim == 96 || query_head_dim == 128);
-
-  const bool supported_vector_config =
-      sdpa_supported_head_dim && query_sequence_length < 4;
-
-  const bool supported_config = supported_vector_config;
-
-  return has_arr_mask || !supported_config;
-}
-
-void ScaledDotProductAttention::eval_gpu(
-    const std::vector<array>& inputs,
-    array& out) {
-  nvtx3::scoped_range r("ScaledDotProductAttention::eval_gpu");
-
-  auto& s = stream();
-  auto& encoder = cu::get_command_encoder(s);
-
-  auto& q_pre = inputs[0];
-  auto& k_pre = inputs[1];
-  auto& v_pre = inputs[2];
-  auto& o = out;
-
-  std::vector<array> copies;
-
-  // Define some copy functions to ensure the layout of the inputs is as
-  // expected.
-  copies.reserve(3);
-  auto copy_unless = [&copies, &s](
-                         auto predicate, const array& arr) -> const array& {
-    if (!predicate(arr)) {
-      array arr_copy = contiguous_copy_gpu(arr, s);
-      copies.push_back(std::move(arr_copy));
-      return copies.back();
-    } else {
-      return arr;
-    }
-  };
-
-  // We are in vector mode ie single query
-  if (q_pre.shape(2) < 4) {
-    auto q_copy_unless = [](const array& arr) {
-      if (arr.flags().row_contiguous) {
-        return true;
-      }
-      auto& strides = arr.strides();
-      auto& shape = arr.shape();
-      if (shape[0] == 1 || shape[1] == 1) {
-        // If either the batch or head dimension is a singleton, the other can
-        // be transposed with the sequence dimension
-        auto bidx = shape[0] == 1 ? 1 : 0;
-        return (strides[3] == 1) && (strides[2] == shape[3] * shape[bidx]) &&
-            (strides[bidx] == shape[3]);
-      }
-      return false;
-    };
-
-    auto kv_copy_unless = [](const array& arr) {
-      // keys and values should be copied if:
-      // - the last dimension is not contiguous
-      // - the batch and head dim are not contiguous
-      auto& strides = arr.strides();
-      auto& shape = arr.shape();
-      if (strides.back() != 1) {
-        return false;
-      }
-      if (shape[0] == 1 || shape[1] == 1) {
-        return true;
-      }
-      return (strides[0] == strides[1] * shape[1]);
-    };
-
-    const auto& q = copy_unless(q_copy_unless, q_pre);
-    const auto& k = copy_unless(kv_copy_unless, k_pre);
-    const auto& v = copy_unless(kv_copy_unless, v_pre);
-
-    for (const auto& cp : copies) {
-      encoder.add_temporary(cp);
-    }
-
-    // Donate the query if possible
-    if (q.is_donatable() && q.flags().row_contiguous && q.size() == o.size()) {
-      o.copy_shared_buffer(q);
-    } else {
-      int64_t str_oD = 1;
-      int64_t str_oH = o.shape(3);
-      int64_t str_oL = o.shape(1) * str_oH;
-      int64_t str_oB = o.shape(2) * str_oL;
-      size_t data_size = o.shape(0) * str_oB;
-
-      array::Flags flags{
-          /* bool contiguous = */ 1,
-          /* bool row_contiguous = */ o.shape(2) == 1,
-          /* bool col_contiguous = */ 0,
-      };
-
-      o.set_data(
-          allocator::malloc(o.nbytes()),
-          data_size,
-          {str_oB, str_oH, str_oL, str_oD},
-          flags);
-    }
-
-    return sdpa_vector_fallback(s, encoder, q, k, v, scale_, o, do_causal_);
-  }
-
-  // Full attention mode should never reach here
-  else {
-    throw std::runtime_error("Doesn't support matrix yet.");
-  }
-}
-
-} // namespace fast
-
-} // namespace mlx::core

mlx/backend/cuda/softmax.cu

@@ -11,6 +11,7 @@
 #include <cooperative_groups.h>
 #include <cooperative_groups/reduce.h>
 #include <nvtx3/nvtx3.hpp>
+#include <cub/block/block_load.cuh>
 
 #include <cassert>
 
@@ -44,21 +45,20 @@ __global__ void softmax(const T* in, T* out, int axis_size) {
   AccT maxval = Limits<AccT>::finite_min();
   AccT normalizer = cast_to<AccT>(0);
   for (int r = 0; r < cuda::ceil_div(axis_size, BLOCK_DIM * N_READS); r++) {
-    auto index = r * BLOCK_DIM + block.thread_rank();
-    auto vals = load_vector<N_READS>(in, index, axis_size, Limits<T>::min());
+    AccT vals[N_READS];
+    cub::LoadDirectBlocked(
+        r * BLOCK_DIM + block.thread_rank(),
+        make_cast_iterator<AccT>(in),
+        vals,
+        axis_size,
+        Limits<AccT>::min());
     prevmax = maxval;
-#pragma unroll
-    for (int i = 0; i < N_READS; ++i) {
-      maxval = max_op(maxval, static_cast<AccT>(vals[i]));
-    }
-
+    maxval = max_op(maxval, cub::ThreadReduce(vals, max_op));
     // Online normalizer calculation for softmax:
     // https://github.com/NVIDIA/online-softmax
     normalizer = normalizer * softmax_exp(prevmax - maxval);
-#pragma unroll
     for (int i = 0; i < N_READS; i++) {
-      normalizer =
-          normalizer + softmax_exp(static_cast<AccT>(vals[i]) - maxval);
+      normalizer = normalizer + softmax_exp(vals[i] - maxval);
     }
   }
 
@@ -95,11 +95,12 @@ __global__ void softmax(const T* in, T* out, int axis_size) {
   // Write output.
   for (int r = 0; r < cuda::ceil_div(axis_size, BLOCK_DIM * N_READS); r++) {
     auto index = r * BLOCK_DIM + block.thread_rank();
-    auto vals = load_vector<N_READS>(in, index, axis_size, T(0));
+    T vals[N_READS];
+    cub::LoadDirectBlocked(index, in, vals, axis_size);
     for (int i = 0; i < N_READS; i++) {
       vals[i] = softmax_exp(static_cast<AccT>(vals[i]) - maxval) * normalizer;
     }
-    store_vector<N_READS>(out, index, vals, axis_size);
+    cub::StoreDirectBlocked(index, out, vals, axis_size);
   }
 }
 
@@ -140,9 +141,9 @@ void Softmax::eval_gpu(const std::vector<array>& inputs, array& out) {
   encoder.set_input_array(in);
   encoder.set_output_array(out);
   dispatch_float_types(out.dtype(), "softmax", [&](auto type_tag) {
-    using DataType = cuda_type_t<MLX_GET_TYPE(type_tag)>;
-    constexpr int N_READS = 16 / sizeof(DataType);
+    constexpr int N_READS = 4;
     dispatch_block_dim(cuda::ceil_div(axis_size, N_READS), [&](auto block_dim) {
+      using DataType = cuda_type_t<MLX_GET_TYPE(type_tag)>;
       auto kernel = cu::softmax<DataType, DataType, block_dim(), N_READS>;
       if (precise) {
         kernel = cu::softmax<DataType, float, block_dim(), N_READS>;

mlx/backend/cuda/sort.cu

@@ -13,6 +13,7 @@
 #include <cub/device/device_segmented_sort.cuh>
 
 #include <cassert>
+#include <numeric>
 
 namespace mlx::core {
 
@@ -26,6 +27,29 @@ struct ModOp {
   }
 };
 
+// We can not use any op in eval, make an utility.
+array swapaxes_in_eval(const array& in, int axis1, int axis2) {
+  std::vector<int> axes(in.ndim());
+  std::iota(axes.begin(), axes.end(), 0);
+  std::swap(axes[axis1], axes[axis2]);
+  // TODO: Share the code with Transpose::eval.
+  Shape shape(axes.size());
+  Strides strides(in.ndim());
+  for (size_t ax = 0; ax < axes.size(); ++ax) {
+    shape[ax] = in.shape()[axes[ax]];
+    strides[ax] = in.strides()[axes[ax]];
+  }
+  auto flags = in.flags();
+  if (flags.contiguous) {
+    auto [_, row_contiguous, col_contiguous] = check_contiguity(shape, strides);
+    flags.row_contiguous = row_contiguous;
+    flags.col_contiguous = col_contiguous;
+  }
+  array out(shape, in.dtype(), nullptr, {});
+  out.copy_shared_buffer(in, strides, flags, in.data_size());
+  return out;
+}
+
 struct OffsetTransform {
   int nsort;
 

mlx/backend/cuda/steel/defines.cuh

@@ -1,9 +0,0 @@
-// Copyright © 2025 Apple Inc.
-
-#pragma once
-
-#define MLX_UNROLL _Pragma("unroll")
-
-#if defined(__CUDA_ARCH__) && (__CUDA_ARCH__ >= 800)
-#define MLX_CUDA_SM_80_ENABLED
-#endif

mlx/backend/cuda/steel/gemm.cuh

@@ -1,101 +0,0 @@
-
-#include "mlx/backend/cuda/steel/mma.cuh"
-#include "mlx/backend/cuda/steel/tiles.cuh"
-
-namespace mlx::core::cu {
-
-/**
- * An example gemm written with the utils.
- *
- * Computes A @ B.T when A and B are all aligned with the block sizes.
- */
-template <typename T, int BM, int BN, int BK>
-__global__ void ab_t_aligned(const T* a, const T* b, T* y, int N, int K) {
-  constexpr int WARPS_M = 2;
-  constexpr int WARPS_N = 2;
-  constexpr int NUM_WARPS = WARPS_M * WARPS_N;
-  constexpr int WARP_STEP_M = BM / WARPS_M;
-  constexpr int WARP_STEP_N = BN / WARPS_N;
-
-  // Precompute some offsets for each thread
-  const int warpid = threadIdx.x / 32;
-  const int laneid = threadIdx.x % 32;
-  const int wm = warpid / WARPS_N;
-  const int wn = warpid % WARPS_N;
-  const int offset_m = wm * WARP_STEP_M;
-  const int offset_n = wn * WARP_STEP_N;
-
-  // Allocate shared memory
-  extern __shared__ char shmem[];
-  SharedTile<T, BM, BK>(&as)[2] = *(SharedTile<T, BM, BK>(*)[2])(&shmem[0]);
-  SharedTile<T, BN, BK>(&bs)[2] =
-      *(SharedTile<T, BN, BK>(*)[2])(&shmem[sizeof(T) * 2 * BM * BK]);
-
-  // Allocate registers for the MMA
-  RegisterTile<float, BM / WARPS_M, BN / WARPS_N> C;
-  RegisterTile<T, BM / WARPS_M, 16> A;
-  RegisterTile<T, BN / WARPS_N, 16> B;
-
-  // Move the global pointers to the tile
-  a += blockIdx.y * BM * K;
-  b += blockIdx.x * BN * K;
-  y += blockIdx.y * BM * N + blockIdx.x * BN;
-
-  // Zero the accumulators
-  C.fill(0);
-
-  // Start the SM pipeline
-  load_async<NUM_WARPS>(as[0], as[0].base_addr(), a, K);
-  load_async<NUM_WARPS>(bs[0], bs[0].base_addr(), b, K);
-  cp_async_commit();
-
-  int tic = 0;
-  for (int k_block = BK; k_block < K; k_block += BK) {
-    load_async<NUM_WARPS>(as[tic ^ 1], as[tic ^ 1].base_addr(), a + k_block, K);
-    load_async<NUM_WARPS>(bs[tic ^ 1], bs[tic ^ 1].base_addr(), b + k_block, K);
-    cp_async_commit();
-    cp_async_wait<1>();
-    __syncthreads();
-
-    MLX_UNROLL
-    for (int k = 0; k < BK / 16; k++) {
-      A.load(
-          as[tic],
-          as[tic].base_addr(),
-          offset_m + laneid % 16,
-          k * 16 + laneid / 16 * 8);
-      B.load(
-          bs[tic],
-          bs[tic].base_addr(),
-          offset_n + laneid % 16,
-          k * 16 + laneid / 16 * 8);
-
-      mma_t(C, A, B);
-    }
-
-    tic ^= 1;
-  }
-
-  // Empty the pipeline
-  cp_async_wait_all();
-  __syncthreads();
-  MLX_UNROLL
-  for (int k = 0; k < BK / 16; k++) {
-    A.load(
-        as[tic],
-        as[tic].base_addr(),
-        offset_m + laneid % 16,
-        k * 16 + laneid / 16 * 8);
-    B.load(
-        bs[tic],
-        bs[tic].base_addr(),
-        offset_n + laneid % 16,
-        k * 16 + laneid / 16 * 8);
-
-    mma_t(C, A, B);
-  }
-
-  C.store_global(y, N, offset_m, offset_n);
-}
-
-} // namespace mlx::core::cu

mlx/backend/cuda/steel/utils.cuh

@@ -1,89 +0,0 @@
-// Copyright © 2025 Apple Inc.
-
-#pragma once
-
-#include "mlx/backend/cuda/device/utils.cuh"
-#include "mlx/backend/cuda/steel/defines.cuh"
-
-namespace mlx::core::cu {
-
-/**
- * Copy bytes from the global memory address pointed to by x to the smem
- * address pointed to by row_address.
- *
- * A simple wrapper over the PTX.
- */
-template <int N, typename T>
-__device__ inline void cp_async(uint32_t row_address, const T* x) {
-  static_assert(
-      N == 16 || N == 8 || N == 4,
-      "cp.async is only supported for N in {4, 8, 16}.");
-#if defined(MLX_CUDA_SM_80_ENABLED)
-  if constexpr (N == 16) {
-    asm volatile(
-        "cp.async.ca.shared::cta.global [%0], [%1], 16;\n" ::"r"(row_address),
-        "l"(reinterpret_cast<const int4*>(x)));
-  } else if constexpr (N == 8) {
-    asm volatile(
-        "cp.async.ca.shared::cta.global [%0], [%1], 8;\n" ::"r"(row_address),
-        "l"(reinterpret_cast<const int2*>(x)));
-  } else if constexpr (N == 4) {
-    asm volatile(
-        "cp.async.ca.shared::cta.global [%0], [%1], 4;\n" ::"r"(row_address),
-        "l"(reinterpret_cast<const int*>(x)));
-  }
-#endif
-}
-
-/**
- * Submit all the previous async copies to be executed.
- */
-__device__ inline void cp_async_commit() {
-#if defined(MLX_CUDA_SM_80_ENABLED)
-  asm volatile("cp.async.commit_group;\n" ::);
-#endif
-}
-
-/**
- * Wait for all but N of the async copies to finish.
- */
-template <int N>
-__device__ inline void cp_async_wait() {
-#if defined(MLX_CUDA_SM_80_ENABLED)
-  if constexpr (N == 0) {
-    asm volatile("cp.async.wait_all;\n" ::);
-  } else {
-    asm volatile("cp.async.wait_group %0;\n" ::"n"(N));
-  }
-#endif
-}
-
-/**
- * Wait for all the async copies to finish.
- */
-__device__ inline void cp_async_wait_all() {
-  cp_async_wait<0>();
-}
-
-/**
- * Extract ``bits`` bits from the 32 bit value.
- *
- * Single instruction shift and mask.
- */
-template <int bits>
-__device__ inline uint32_t extract_bits(uint32_t value, int start_bit) {
-  static_assert(
-      bits == 2 || bits == 4 || bits == 8,
-      "extract_bits only supports 2, 4, 8 for now.");
-  uint32_t result;
-  if constexpr (bits == 2) {
-    asm("bfe.u32 %0, %1, %2, 2;" : "=r"(result) : "r"(value), "r"(start_bit));
-  } else if constexpr (bits == 4) {
-    asm("bfe.u32 %0, %1, %2, 4;" : "=r"(result) : "r"(value), "r"(start_bit));
-  } else if constexpr (bits == 8) {
-    asm("bfe.u32 %0, %1, %2, 8;" : "=r"(result) : "r"(value), "r"(start_bit));
-  }
-  return result;
-}
-
-} // namespace mlx::core::cu

mlx/backend/cuda/ternary.cu

@@ -32,7 +32,7 @@ ternary_v(const bool* a, const T* b, const T* c, T* out, IdxT size) {
     AlignedVector<T, N_READS> out_vec;
 #pragma unroll
     for (int i = 0; i < N_READS; ++i) {
-      out_vec[i] = Op{}(a_vec[i], b_vec[i], c_vec[i]);
+      out_vec.val[i] = Op{}(a_vec.val[i], b_vec.val[i], c_vec.val[i]);
     }
 
     store_vector<N_READS>(out, index, out_vec);
@@ -76,7 +76,7 @@ __global__ void ternary_g(
     int ndim) {
   IdxT index = cg::this_grid().thread_rank();
   if (index < size) {
-    auto [a_idx, b_idx, c_idx] = elem_to_loc(
+    auto [a_idx, b_idx, c_idx] = elem_to_loc_4d(
         index,
         shape.data(),
         a_strides.data(),
@@ -125,9 +125,12 @@ void ternary_op_gpu_inplace(
             int ndim = shape.size();
             if (ndim <= 3) {
               dispatch_1_2_3(ndim, [&](auto dims_constant) {
-                auto [num_blocks, block_dims] = get_launch_args(out, large());
+                auto kernel =
+                    cu::ternary_g_nd<Op, DType, IdxT, dims_constant()>;
+                auto [num_blocks, block_dims] =
+                    get_launch_args(kernel, out, large());
                 encoder.add_kernel_node(
-                    cu::ternary_g_nd<Op, DType, IdxT, dims_constant()>,
+                    kernel,
                     num_blocks,
                     block_dims,
                     0,
@@ -142,9 +145,11 @@ void ternary_op_gpu_inplace(
                     const_param<dims_constant()>(c_strides));
               });
             } else {
-              auto [num_blocks, block_dims] = get_launch_args(out, large());
+              auto kernel = cu::ternary_g<Op, DType, IdxT>;
+              auto [num_blocks, block_dims] =
+                  get_launch_args(kernel, out, large());
               encoder.add_kernel_node(
-                  cu::ternary_g<Op, DType, IdxT>,
+                  kernel,
                   num_blocks,
                   block_dims,
                   0,
@@ -163,11 +168,18 @@ void ternary_op_gpu_inplace(
     } else {
       dispatch_bool(out.data_size() > UINT32_MAX, [&](auto large) {
         using IdxT = std::conditional_t<large(), int64_t, uint32_t>;
-        constexpr int N_READS = 16 / sizeof(DType);
+        // TODO: Choose optimized value based on type size.
+        constexpr int N_READS = 4;
+        auto kernel = cu::ternary_v<Op, DType, IdxT, N_READS>;
         auto [num_blocks, block_dims] = get_launch_args(
-            out.data_size(), out.shape(), out.strides(), large(), N_READS);
+            kernel,
+            out.data_size(),
+            out.shape(),
+            out.strides(),
+            large(),
+            N_READS);
         encoder.add_kernel_node(
-            cu::ternary_v<Op, DType, IdxT, N_READS>,
+            kernel,
             num_blocks,
             block_dims,
             0,
@@ -195,7 +207,7 @@ void ternary_op_gpu(
 }
 
 void Select::eval_gpu(const std::vector<array>& inputs, array& out) {
-  nvtx3::scoped_range r("Select::eval_gpu");
+  nvtx3::scoped_range r("select::eval_gpu");
   auto& s = out.primitive().stream();
   ternary_op_gpu<cu::Select>(inputs, out, s);
 }

mlx/backend/cuda/unary.cu

@@ -3,6 +3,7 @@
 #include "mlx/backend/common/unary.h"
 #include "mlx/backend/cuda/device.h"
 #include "mlx/backend/cuda/device/unary_ops.cuh"
+#include "mlx/backend/cuda/iterators/general_iterator.cuh"
 #include "mlx/backend/cuda/kernel_utils.cuh"
 #include "mlx/dtype_utils.h"
 #include "mlx/primitives.h"
@@ -30,7 +31,7 @@ __global__ void unary_v(const In* in, Out* out, IdxT size) {
     AlignedVector<Out, N_READS> out_vec;
 #pragma unroll
     for (int i = 0; i < N_READS; ++i) {
-      out_vec[i] = Op{}(in_vec[i]);
+      out_vec.val[i] = Op{}(in_vec.val[i]);
     }
 
     store_vector<N_READS>(out, index, out_vec);
@@ -47,7 +48,7 @@ __global__ void unary_g(
     int ndim) {
   IdxT index = cg::this_grid().thread_rank();
   if (index < size) {
-    auto idx = elem_to_loc(index, shape.data(), strides.data(), ndim);
+    auto idx = elem_to_loc_4d(index, shape.data(), strides.data(), ndim);
     out[index] = Op{}(in[idx]);
   }
 }
@@ -129,10 +130,16 @@ void unary_op_gpu_inplace(
             using IdxT = std::conditional_t<large(), int64_t, uint32_t>;
             // TODO: Choose optimized value based on type size.
             constexpr int N_READS = 4;
+            auto kernel = cu::unary_v<Op, InType, OutType, IdxT, N_READS>;
             auto [num_blocks, block_dims] = get_launch_args(
-                out.data_size(), out.shape(), out.strides(), large, N_READS);
+                kernel,
+                out.data_size(),
+                out.shape(),
+                out.strides(),
+                large,
+                N_READS);
             encoder.add_kernel_node(
-                cu::unary_v<Op, InType, OutType, IdxT, N_READS>,
+                kernel,
                 num_blocks,
                 block_dims,
                 0,
@@ -142,9 +149,10 @@ void unary_op_gpu_inplace(
           } else {
             using IdxT = std::conditional_t<large(), int64_t, int32_t>;
             auto [shape, strides] = collapse_contiguous_dims(in);
-            auto [num_blocks, block_dims] = get_launch_args(out, large);
+            auto kernel = cu::unary_g<Op, InType, OutType, IdxT>;
+            auto [num_blocks, block_dims] = get_launch_args(kernel, out, large);
             encoder.add_kernel_node(
-                cu::unary_g<Op, InType, OutType, IdxT>,
+                kernel,
                 num_blocks,
                 block_dims,
                 0,

mlx/backend/cuda/utils.cpp

@@ -17,35 +17,6 @@ CudaStream::~CudaStream() {
   CHECK_CUDA_ERROR(cudaStreamDestroy(stream_));
 }
 
-CudaGraphExec::CudaGraphExec(cudaGraphExec_t handle) : handle_(handle) {}
-
-CudaGraphExec::CudaGraphExec(CudaGraphExec&& other) : handle_(other.handle_) {
-  other.handle_ = nullptr;
-};
-
-CudaGraphExec::~CudaGraphExec() {
-  reset();
-}
-
-void CudaGraphExec::instantiate(cudaGraph_t graph) {
-  CHECK_CUDA_ERROR(cudaGraphInstantiate(&handle_, graph, nullptr, nullptr, 0));
-}
-
-void CudaGraphExec::reset() {
-  if (handle_ != nullptr) {
-    CHECK_CUDA_ERROR(cudaGraphExecDestroy(handle_));
-    handle_ = nullptr;
-  }
-}
-
-void check_cublas_error(const char* name, cublasStatus_t err) {
-  if (err != CUBLAS_STATUS_SUCCESS) {
-    // TODO: Use cublasGetStatusString when it is widely available.
-    throw std::runtime_error(
-        fmt::format("{} failed with code: {}.", name, static_cast<int>(err)));
-  }
-}
-
 void check_cuda_error(const char* name, cudaError_t err) {
   if (err != cudaSuccess) {
     throw std::runtime_error(

mlx/backend/cuda/utils.h

@@ -4,7 +4,6 @@
 
 #pragma once
 
-#include <cublasLt.h>
 #include <cuda.h>
 #include <cuda_runtime.h>
 
@@ -33,34 +32,11 @@ class CudaStream {
   cudaStream_t stream_;
 };
 
-// Move-able RAII handle of cudaGraphExec_t.
-class CudaGraphExec {
- public:
-  CudaGraphExec(cudaGraphExec_t handle = nullptr);
-  CudaGraphExec(CudaGraphExec&& other);
-  ~CudaGraphExec();
-
-  CudaGraphExec(const CudaGraphExec&) = delete;
-  CudaGraphExec& operator=(const CudaGraphExec&) = delete;
-
-  void instantiate(cudaGraph_t graph);
-  void reset();
-
-  operator cudaGraphExec_t() const {
-    return handle_;
-  }
-
- private:
-  cudaGraphExec_t handle_;
-};
-
 // Throw exception if the cuda API does not succeed.
-void check_cublas_error(const char* name, cublasStatus_t err);
 void check_cuda_error(const char* name, cudaError_t err);
 void check_cuda_error(const char* name, CUresult err);
 
 // The macro version that prints the command that failed.
-#define CHECK_CUBLAS_ERROR(cmd) check_cublas_error(#cmd, (cmd))
 #define CHECK_CUDA_ERROR(cmd) check_cuda_error(#cmd, (cmd))
 
 // Convert Dtype to CUDA C++ types.

mlx/backend/cuda/worker.cpp

@@ -1,6 +1,7 @@
 // Copyright © 2025 Apple Inc.
 
 #include "mlx/backend/cuda/worker.h"
+#include "mlx/backend/cuda/allocator.h"
 #include "mlx/backend/cuda/device.h"
 
 namespace mlx::core::cu {
@@ -11,10 +12,10 @@ Worker::Worker()
 
 Worker::~Worker() {
   {
-    std::lock_guard lock(mtx_);
+    std::lock_guard lock(worker_mutex_);
     stop_ = true;
   }
-  cond_.notify_one();
+  worker_event_.signal(batch_ + 1);
   worker_.join();
 }
 
@@ -22,41 +23,53 @@ void Worker::add_task(std::function<void()> task) {
   pending_tasks_.push_back(std::move(task));
 }
 
-void Worker::signal(void* data) {
-  auto w = static_cast<Worker*>(data);
-  {
-    std::lock_guard lock(w->mtx_);
-    w->signaled_batch_++;
+void Worker::consume_in_this_thread() {
+  for (auto& task : pending_tasks_) {
+    task();
   }
-  w->cond_.notify_one();
+  pending_tasks_.clear();
+}
+
+void Worker::end_batch() {
+  batch_++;
+  {
+    std::lock_guard lock(worker_mutex_);
+    worker_tasks_[batch_] = std::move(pending_tasks_);
+  }
+  uncommited_batches_++;
+}
+
+void Worker::commit() {
+  if (uncommited_batches_ == 0) {
+    return;
+  }
+  uncommited_batches_ = 0;
+  worker_event_.signal(batch_);
 }
 
 void Worker::commit(cudaStream_t stream) {
-  // Move pending tasks into tasks
-  if (pending_tasks_.empty()) {
+  if (uncommited_batches_ == 0) {
     return;
   }
-  {
-    std::lock_guard lock(mtx_);
-    // Move pending tasks into ready tasks
-    worker_tasks_[++committed_batch_] = std::move(pending_tasks_);
-  }
+  uncommited_batches_ = 0;
+  // Signal the |worker_event_| in |signal_stream_| after the kernels in
+  // |stream_| finish running.
   signal_event_.record(stream);
   signal_event_.wait(signal_stream_);
-  cudaLaunchHostFunc(signal_stream_, signal, this);
+  worker_event_.signal(signal_stream_, batch_);
 }
 
 void Worker::thread_fn() {
+  // The worker thread is safe to free buffers.
+  allocator().register_this_thread();
+
   while (!stop_) {
-    uint64_t current_batch = 0;
+    uint64_t batch = worker_event_.value();
     Tasks tasks;
     {
-      std::unique_lock<std::mutex> lk(mtx_);
-      cond_.wait(lk, [this, &current_batch] {
-        return this->signaled_batch_ > current_batch || this->stop_;
-      });
-      current_batch = signaled_batch_;
-      auto end = worker_tasks_.upper_bound(current_batch);
+      std::lock_guard lock(worker_mutex_);
+      // Move tasks in signaled batches.
+      auto end = worker_tasks_.upper_bound(batch);
       for (auto it = worker_tasks_.begin(); it != end; ++it) {
         if (tasks.empty()) {
           tasks = std::move(it->second);
@@ -72,6 +85,7 @@ void Worker::thread_fn() {
       auto task = std::move(tasks[i]);
       task();
     }
+    worker_event_.wait(batch + 1);
   }
 }
 

mlx/backend/cuda/worker.h

@@ -5,7 +5,6 @@
 #include "mlx/backend/cuda/event.h"
 #include "mlx/backend/cuda/utils.h"
 
-#include <condition_variable>
 #include <functional>
 #include <map>
 #include <mutex>
@@ -25,24 +24,38 @@ class Worker {
   // Add a pending |task| that will run when consumed or commited.
   void add_task(std::function<void()> task);
 
+  // Run pending tasks immediately in current thread.
+  void consume_in_this_thread();
+
+  // Put pending tasks in a batch.
+  void end_batch();
+
+  // Inform worker thread to run current batches now.
+  void commit();
+
   // Inform worker thread to run current batches after kernels in |stream|
   // finish running.
   void commit(cudaStream_t stream);
 
+  // Return how many batches have been added but not committed yet.
+  size_t uncommited_batches() const {
+    return uncommited_batches_;
+  }
+
  private:
-  static void signal(void*);
-
   void thread_fn();
-  std::mutex mtx_;
-  std::condition_variable cond_;
 
-  uint64_t committed_batch_{0};
-  uint64_t signaled_batch_{0};
+  uint64_t batch_{0};
+  size_t uncommited_batches_{0};
 
   // Cuda stream and event for signaling kernel completion.
   CudaStream signal_stream_;
   CudaEvent signal_event_;
 
+  // Worker thread.
+  SharedEvent worker_event_;
+  std::thread worker_;
+  std::mutex worker_mutex_;
   bool stop_{false};
 
   // Tasks are put in |pending_tasks_| first, and then moved to
@@ -50,7 +63,6 @@ class Worker {
   using Tasks = std::vector<std::function<void()>>;
   Tasks pending_tasks_;
   std::map<uint64_t, Tasks> worker_tasks_;
-  std::thread worker_;
 };
 
 } // namespace mlx::core::cu

mlx/backend/gpu/primitives.cpp

@@ -133,7 +133,6 @@ void NumberOfElements::eval_gpu(const std::vector<array>& inputs, array& out) {
 }
 
 void Pad::eval_gpu(const std::vector<array>& inputs, array& out) {
-  MLX_PROFILER_RANGE("Pad::eval_gpu");
   // Inputs must be base input array and scalar val array
   assert(inputs.size() == 2);
   auto& in = inputs[0];

mlx/backend/metal/allocator.cpp

@@ -128,7 +128,8 @@ Buffer MetalAllocator::malloc(size_t size) {
 
     auto pool = metal::new_scoped_memory_pool();
 
-    // If we have a lot of memory pressure try to reclaim memory from the cache
+    // If we have a lot of memory pressure or are over the maximum cache size,
+    // try to reclaim memory from the cache
     if (mem_required >= gc_limit_ || num_resources_ >= resource_limit_) {
       num_resources_ -=
           buffer_cache_.release_cached_buffers(mem_required - gc_limit_);

mlx/backend/metal/device.h

@@ -60,16 +60,6 @@ struct CommandEncoder {
     enc_->updateFence(fence);
   }
 
-  template <typename T>
-  void set_vector_bytes(const SmallVector<T>& vec, size_t nelems, int idx) {
-    enc_->setBytes(vec.data(), nelems * sizeof(T), idx);
-  }
-  template <typename T>
-  void set_vector_bytes(const SmallVector<T>& vec, int idx) {
-    return set_vector_bytes(vec, vec.size(), idx);
-  }
-
-  // TODO: Code is duplicated but they should be deleted soon.
   template <typename T>
   void set_vector_bytes(const std::vector<T>& vec, size_t nelems, int idx) {
     enc_->setBytes(vec.data(), nelems * sizeof(T), idx);
@@ -104,7 +94,7 @@ struct CommandEncoder {
   };
 
   // Outputs of all kernels in the encoder including temporaries
-  std::unordered_set<const void*>& outputs() {
+  std::unordered_set<const void*> outputs() {
     return all_outputs_;
   };
 

mlx/backend/metal/event.cpp

@@ -14,10 +14,6 @@ Event::Event(Stream stream) : stream_(stream) {
   auto p = metal::new_scoped_memory_pool();
   event_ = std::shared_ptr<void>(
       metal::device(Device::gpu).mtl_device()->newSharedEvent(), dtor);
-  if (event_ == nullptr) {
-    throw std::runtime_error(
-        "[Event::Event] Failed to create Metal shared event.");
-  }
 }
 
 void Event::wait() {

mlx/backend/metal/quantized.cpp

@@ -32,20 +32,15 @@ inline array ensure_row_contiguous_matrix(
     const array& x,
     metal::Device& d,
     const Stream& s) {
-  if (x.ndim() < 2) {
-    if (x.strides()[0] == 1) {
-      return x;
-    }
+  auto stride_0 = x.strides()[x.ndim() - 2];
+  auto stride_1 = x.strides()[x.ndim() - 1];
+  if (stride_0 == x.shape(-1) && stride_1 == 1) {
+    return x;
   } else {
-    auto stride_0 = x.strides()[x.ndim() - 2];
-    auto stride_1 = x.strides()[x.ndim() - 1];
-    if (stride_0 == x.shape(-1) && stride_1 == 1) {
-      return x;
-    }
+    array x_copy = contiguous_copy_gpu(x, s);
+    d.add_temporary(x_copy, s.index);
+    return x_copy;
   }
-  array x_copy = contiguous_copy_gpu(x, s);
-  d.add_temporary(x_copy, s.index);
-  return x_copy;
 }
 
 inline int get_qmv_batch_limit(int D, int O, metal::Device& d) {
@@ -270,15 +265,9 @@ void qvm_split_k(
   MTL::Size group_dims = MTL::Size(bk, 2, 1);
   MTL::Size grid_dims = MTL::Size(M, N / bn, B);
 
+  int x_batch_ndims = x.ndim() - 2;
   auto x_shape = x.shape();
   auto x_strides = x.strides();
-  if (x_shape.size() == 1) {
-    x_shape.insert(x_shape.begin(), 1);
-    x_strides.insert(x_strides.begin(), 0);
-  }
-
-  int x_ndim = x_shape.size();
-  int x_batch_ndims = x_ndim - 2;
   int w_batch_ndims = w.ndim() - 2;
   auto w_shape = w.shape();
   auto w_strides = w.strides();
@@ -289,7 +278,7 @@ void qvm_split_k(
   x_shape.insert(x_shape.end() - 2, split_k);
   x_shape.back() /= split_k;
   x_strides.insert(x_strides.end() - 2, split_D);
-  x_strides[x_ndim - 1] = split_D;
+  x_strides[x.ndim() - 1] = split_D;
   x_batch_ndims += 1;
 
   w_shape.insert(w_shape.end() - 2, split_k);
@@ -302,9 +291,6 @@ void qvm_split_k(
   int final_block_size = K - (split_k - 1) * split_D;
 
   auto temp_shape = out.shape();
-  if (temp_shape.size() == 1) {
-    temp_shape.insert(temp_shape.begin(), 1);
-  }
   temp_shape.insert(temp_shape.end() - 2, split_k);
   array intermediate(temp_shape, x.dtype(), nullptr, {});
   intermediate.set_data(allocator::malloc(intermediate.nbytes()));

mlx/distributed/CMakeLists.txt

@@ -6,4 +6,3 @@ target_sources(
 
 add_subdirectory(${CMAKE_CURRENT_SOURCE_DIR}/mpi)
 add_subdirectory(${CMAKE_CURRENT_SOURCE_DIR}/ring)
-add_subdirectory(${CMAKE_CURRENT_SOURCE_DIR}/nccl)

mlx/distributed/distributed.cpp

@@ -2,11 +2,9 @@
 
 #include <unordered_map>
 
-#include <iostream>
 #include "mlx/distributed/distributed.h"
 #include "mlx/distributed/distributed_impl.h"
 #include "mlx/distributed/mpi/mpi.h"
-#include "mlx/distributed/nccl/nccl.h"
 #include "mlx/distributed/ring/ring.h"
 
 namespace mlx::core::distributed {
@@ -82,7 +80,7 @@ class EmptyGroup : public GroupImpl {
 } // namespace detail
 
 bool is_available() {
-  return mpi::is_available() || ring::is_available() || nccl::is_available();
+  return mpi::is_available() || ring::is_available();
 }
 
 int Group::rank() const {
@@ -113,8 +111,6 @@ Group init(bool strict /* = false */, const std::string& bk /* = "any" */) {
     group = mpi::init(strict);
   } else if (bk == "ring") {
     group = ring::init(strict);
-  } else if (bk == "nccl") {
-    group = nccl::init(strict);
   } else if (bk == "any") {
     group = ring::init(false);
     bk_ = "ring";

mlx/distributed/distributed.h

@@ -3,6 +3,7 @@
 #pragma once
 
 #include <memory>
+
 #include "mlx/array.h"
 
 namespace mlx::core::distributed {

mlx/distributed/nccl/CMakeLists.txt

@@ -1,8 +0,0 @@
-if(MLX_BUILD_CUDA)
-  target_sources(mlx PRIVATE ${CMAKE_CURRENT_SOURCE_DIR}/nccl.cpp)
-  find_package(NCCL REQUIRED)
-  target_link_libraries(mlx PRIVATE ${NCCL_LIBRARIES})
-  target_include_directories(mlx PRIVATE ${NCCL_INCLUDE_DIRS})
-else()
-  target_sources(mlx PRIVATE ${CMAKE_CURRENT_SOURCE_DIR}/no_nccl.cpp)
-endif()

mlx/distributed/nccl/nccl.cpp

@@ -1,359 +0,0 @@
-#include <arpa/inet.h>
-#include <cuda_runtime.h>
-#include <nccl.h>
-#include <netdb.h>
-#include <sys/socket.h>
-#include <unistd.h>
-#include <cstdio>
-#include <cstdlib>
-#include <cstring>
-#include <iostream>
-#include <mutex>
-#include <stdexcept>
-#include <string>
-#include <type_traits>
-
-#include "mlx/backend/cuda/device.h"
-#include "mlx/distributed/distributed.h"
-#include "mlx/distributed/distributed_impl.h"
-
-namespace mlx::core::distributed::nccl {
-
-#define CHECK_CUDA(cmd)              \
-  do {                               \
-    cudaError_t e = cmd;             \
-    if (e != cudaSuccess) {          \
-      fprintf(                       \
-          stderr,                    \
-          "CUDA error %s:%d '%s'\n", \
-          __FILE__,                  \
-          __LINE__,                  \
-          cudaGetErrorString(e));    \
-      exit(1);                       \
-    }                                \
-  } while (0)
-
-#define CHECK_NCCL(cmd)              \
-  do {                               \
-    ncclResult_t r = cmd;            \
-    if (r != ncclSuccess) {          \
-      fprintf(                       \
-          stderr,                    \
-          "NCCL error %s:%d '%s'\n", \
-          __FILE__,                  \
-          __LINE__,                  \
-          ncclGetErrorString(r));    \
-      exit(1);                       \
-    }                                \
-  } while (0)
-
-namespace detail {
-
-inline void sendAll(int sock, const void* buf, size_t len) {
-  const char* ptr = reinterpret_cast<const char*>(buf);
-  while (len > 0) {
-    ssize_t sent = send(sock, ptr, len, 0);
-    if (sent <= 0) {
-      perror("send");
-      exit(1);
-    }
-    ptr += sent;
-    len -= sent;
-  }
-}
-
-inline void recvAll(int sock, void* buf, size_t len) {
-  char* ptr = reinterpret_cast<char*>(buf);
-  while (len > 0) {
-    ssize_t rec = recv(sock, ptr, len, 0);
-    if (rec <= 0) {
-      perror("recv");
-      exit(1);
-    }
-    ptr += rec;
-    len -= rec;
-  }
-}
-
-inline void bootstrap_unique_id(
-    ncclUniqueId& id,
-    int rank,
-    int size,
-    const std::string& initMethod) {
-  // Parse the init method to extract the host and port
-  if (initMethod.rfind("tcp://", 0) != 0)
-    throw;
-  auto hostport = initMethod.substr(6);
-  auto colon = hostport.find(':');
-  std::string host = hostport.substr(0, colon);
-  int port = std::stoi(hostport.substr(colon + 1));
-
-  if (rank == 0) {
-    // create a unique id on the rank 0
-    CHECK_NCCL(ncclGetUniqueId(&id));
-
-    // create a socket to send the unique id to all other ranks
-    int sock = socket(AF_INET, SOCK_STREAM, 0);
-
-    if (sock < 0) {
-      std::ostringstream msg;
-      msg << "[nccl] Couldn't create socket (error: " << errno << ")";
-      throw std::runtime_error(msg.str());
-    }
-
-    sockaddr_in serv = {};
-    serv.sin_family = AF_INET;
-    serv.sin_addr.s_addr = htonl(INADDR_ANY);
-    serv.sin_port = htons(port);
-
-    int reuse = 1;
-    // Without this, if rank-0 crashes or restarts process quickly,
-    // the OS might refuse to let binding to the same port, so reuse
-
-    if (setsockopt(sock, SOL_SOCKET, SO_REUSEADDR, &reuse, sizeof(reuse)) < 0) {
-      std::ostringstream msg;
-      msg << "[nccl] setsockopt() failed: " << strerror(errno);
-      throw std::runtime_error(msg.str());
-    }
-
-    if (bind(sock, reinterpret_cast<sockaddr*>(&serv), sizeof(serv)) < 0) {
-      std::ostringstream msg;
-      msg << "[nccl] bind() failed: " << strerror(errno);
-      throw std::runtime_error(msg.str());
-    }
-    if (listen(sock, size - 1) < 0) {
-      std::ostringstream msg;
-      msg << "[nccl] listen() failed: " << strerror(errno);
-      throw std::runtime_error(msg.str());
-    }
-
-    for (int peer = 1; peer < size; ++peer) {
-      int conn = accept(sock, nullptr, nullptr);
-      if (conn < 0) {
-        std::ostringstream msg;
-        msg << "[nccl] accept() failed: " << strerror(errno);
-        throw std::runtime_error(msg.str());
-      }
-      sendAll(conn, &id, sizeof(id));
-      close(conn);
-    }
-    close(sock);
-
-  } else {
-    // Here just wanted to make show that rank 0 has enough time to bind
-    // so we will retry to connect until max attempts
-
-    int sock = socket(AF_INET, SOCK_STREAM, 0);
-    if (sock < 0) {
-      std::ostringstream msg;
-      msg << "[nccl] socket() failed: " << strerror(errno);
-      throw std::runtime_error(msg.str());
-    }
-
-    hostent* he = gethostbyname(host.c_str());
-    if (!he) {
-      throw std::runtime_error("[nccl] lookup failed for host: " + host);
-    }
-    sockaddr_in serv = {};
-    serv.sin_family = AF_INET;
-    memcpy(&serv.sin_addr, he->h_addr_list[0], he->h_length);
-    serv.sin_port = htons(port);
-
-    const int max_retries = 30;
-    int attempt = 0;
-    bool connected = false;
-
-    for (attempt = 0; attempt < max_retries; ++attempt) {
-      if (connect(sock, reinterpret_cast<sockaddr*>(&serv), sizeof(serv)) ==
-          0) {
-        connected = true;
-        std::cout << "[Rank " << rank << "] Connected successfully on attempt "
-                  << attempt + 1 << std::endl;
-        break;
-      }
-      if (errno != ECONNREFUSED) {
-        break;
-      }
-      std::this_thread::sleep_for(std::chrono::milliseconds(500));
-    }
-
-    if (!connected) {
-      std::ostringstream msg;
-      msg << "[Rank " << rank << "] connect() failed after " << attempt
-          << " retries: " << strerror(errno);
-      close(sock);
-      throw std::runtime_error(msg.str());
-    }
-    recvAll(sock, &id, sizeof(id));
-    close(sock);
-  }
-}
-
-template <typename T>
-struct type_identity {
-  using type = T;
-};
-
-template <typename F>
-void dispatch_dtype(const array& arr, F&& f) {
-  switch (arr.dtype()) {
-    case bool_:
-      throw std::invalid_argument("[nccl] Boolean arrays not supported");
-    case int8:
-      f(type_identity<int8_t>{}, ncclChar);
-      break;
-    case uint8:
-      f(type_identity<uint8_t>{}, ncclUint8);
-      break;
-    case int32:
-      f(type_identity<int32_t>{}, ncclInt);
-      break;
-    case uint32:
-      f(type_identity<uint32_t>{}, ncclUint32);
-      break;
-    case int64:
-      f(type_identity<int64_t>{}, ncclInt64);
-      break;
-    case uint64:
-      f(type_identity<uint64_t>{}, ncclUint64);
-      break;
-    case float16:
-      f(type_identity<float16_t>{}, ncclHalf);
-      break;
-    case bfloat16:
-      f(type_identity<bfloat16_t>{}, ncclBfloat16);
-      break;
-    case float32:
-      f(type_identity<float>{}, ncclFloat);
-      break;
-    case float64:
-      f(type_identity<double>{}, ncclDouble);
-      break;
-    default:
-      throw std::invalid_argument("[nccl] Unknown or unsupported dtype");
-  }
-}
-
-} // namespace detail
-
-using GroupImpl = mlx::core::distributed::detail::GroupImpl;
-class NCCLGroup : public GroupImpl {
- public:
-  NCCLGroup(int worldRank, int worldSize, const std::string initMethod)
-      : rank_(worldRank),
-        size_(worldSize),
-        comm_(nullptr),
-        initMethod_(initMethod) {
-    if (initialized_)
-      return;
-    int ndev;
-    CHECK_CUDA(cudaGetDeviceCount(&ndev));
-    CHECK_CUDA(cudaSetDevice(rank_ % ndev));
-    detail::bootstrap_unique_id(uniqueId_, rank_, size_, initMethod_);
-    CHECK_NCCL(ncclCommInitRank(&comm_, size_, uniqueId_, rank_));
-    initialized_ = true;
-  }
-
-  ~NCCLGroup() {
-    ncclCommDestroy(comm_);
-    ncclGroupEnd();
-    initialized_ = false;
-  }
-
-  int rank() override {
-    return rank_;
-  }
-
-  int size() override {
-    return size_;
-  }
-
-  void all_sum(const array& input, array& output, Stream stream) override {
-    detail::dispatch_dtype(input, [&](auto type_tag, ncclDataType_t dt) {
-      using T = typename decltype(type_tag)::type;
-      all_reduce_impl<T>(input, output, stream, dt, ncclSum);
-    });
-  }
-
-  virtual std::shared_ptr<GroupImpl> split(int color, int key = -1) override {
-    throw std::runtime_error("[nccl] Group split not supported.");
-  }
-
-  void all_gather(const array& input, array& output, Stream stream) override {
-    throw std::runtime_error(
-        "[nccl] All gather not supported in NCCL backend.");
-  }
-
-  void send(const array& input, int dst, Stream stream) override {
-    throw std::runtime_error("[nccl] Send not supported in NCCL backend.");
-  }
-
-  void recv(array& output, int src, Stream stream) override {
-    throw std::runtime_error("[nccl] Recv not supported in NCCL backend.");
-  }
-
-  void all_max(const array& input, array& output, Stream stream) override {
-    throw std::runtime_error("[nccl] All max not supported in NCCL backend.");
-  }
-
-  void all_min(const array& input, array& output, Stream stream) override {
-    throw std::runtime_error("[nccl] All min not supported in NCCL backend.");
-  }
-
-  template <typename T>
-  void all_reduce_impl(
-      const array& input,
-      array& output,
-      Stream stream,
-      ncclDataType_t dt,
-      ncclRedOp_t op) {
-    auto& encoder = cu::get_command_encoder(stream);
-
-    CHECK_NCCL(ncclAllReduce(
-        input.data<T>(),
-        output.data<T>(),
-        input.size(),
-        dt,
-        op,
-        comm_,
-        encoder.stream()));
-  }
-
-  int rank_, size_;
-  std::string initMethod_;
-  ncclUniqueId uniqueId_;
-  ncclComm_t comm_;
-  bool initialized_ = false;
-};
-
-bool is_available() {
-  return true;
-}
-
-namespace detail {
-static std::string get_env_var_or_throw(const char* env_var_name) {
-  const char* value = std::getenv(env_var_name);
-  if (value == nullptr) {
-    std::ostringstream msg;
-    msg << "[nccl] Required environment variable '" << env_var_name
-        << "' is not set. "
-        << "Please set it before initializing the distributed backend.";
-    throw std::runtime_error(msg.str());
-  }
-  return std::string(value);
-}
-} // namespace detail
-
-std::shared_ptr<GroupImpl> init(bool strict /* = false */) {
-  std::string host = detail::get_env_var_or_throw("NCCL_HOST_IP");
-  std::string port = detail::get_env_var_or_throw("NCCL_PORT");
-  std::string rank_str = detail::get_env_var_or_throw("MLX_RANK");
-  std::string n_nodes_str = detail::get_env_var_or_throw("MLX_WORLD_SIZE");
-
-  int rank = std::stoi(rank_str);
-  int n_nodes = std::stoi(n_nodes_str);
-  std::string init_method = "tcp://" + host + ":" + port;
-
-  return std::make_shared<NCCLGroup>(rank, n_nodes, init_method);
-}
-} // namespace mlx::core::distributed::nccl

mlx/distributed/nccl/nccl.h

@@ -1,12 +0,0 @@
-// Copyright © 2024 Apple Inc.
-
-#include "mlx/distributed/distributed.h"
-
-namespace mlx::core::distributed::nccl {
-
-using GroupImpl = mlx::core::distributed::detail::GroupImpl;
-
-bool is_available();
-std::shared_ptr<GroupImpl> init(bool strict = false);
-
-} // namespace mlx::core::distributed::nccl

mlx/distributed/nccl/no_nccl.cpp

@@ -1,20 +0,0 @@
-// Copyright © 2024 Apple Inc.
-
-#include "mlx/distributed/nccl/nccl.h"
-
-namespace mlx::core::distributed::nccl {
-
-using GroupImpl = mlx::core::distributed::detail::GroupImpl;
-
-bool is_available() {
-  return false;
-}
-
-std::shared_ptr<GroupImpl> init(bool strict /* = false */) {
-  if (strict) {
-    throw std::runtime_error("Cannot initialize nccl distributed backend.");
-  }
-  return nullptr;
-}
-
-} // namespace mlx::core::distributed::nccl

mlx/distributed/ops.cpp

@@ -31,7 +31,8 @@ array all_sum(
   return array(
       x.shape(),
       x.dtype(),
-      std::make_shared<AllReduce>(to_stream(s), group, AllReduce::Sum),
+      std::make_shared<AllReduce>(
+          to_stream(s, Device::cpu), group, AllReduce::Sum),
       {x});
 }
 

mlx/distributed/ring/ring.cpp

@@ -975,6 +975,7 @@ class RingGroup : public GroupImpl {
 
   int rank_;
   int size_;
+
   bool verbose_;
 
   ThreadPool pool_;