IREElink

IREE (Intermediate Representation Execution Environment¹) is an MLIR-based end-to-end compiler and runtime that lowers Machine Learning (ML) models to a unified IR that scales up to meet the needs of the datacenter and down to satisfy the constraints and special considerations of mobile and edge deployments.

Key featureslink

Ahead-of-time compilation

Scheduling and execution logic are compiled together

Project architecture
Support for advanced model features

Dynamic shapes, flow control, streaming, and more

Importing from ML frameworks
Designed for CPUs, GPUs, and other accelerators

First class support for many popular devices and APIs

Deployment configurations
Low overhead, pipelined execution

Efficient power and resource usage on server and edge devices

Benchmarking
Binary size as low as 30KB on embedded systems

Running on bare-metal
Debugging and profiling support

Profiling with Tracy

Support matrixlink

IREE supports importing from a variety of ML frameworks:

JAX
ONNX
PyTorch
TensorFlow and TensorFlow Lite

The IREE compiler tools run on Linux, Windows, and macOS and can generate efficient code for a variety of runtime platforms:

and architectures:

ARM
x86
RISC-V

Support for hardware accelerators and APIs is also included:

Project architecturelink

IREE adopts a holistic approach towards ML model compilation: the IR produced contains both the scheduling logic, required to communicate data dependencies to low-level parallel pipelined hardware/API like Vulkan, and the execution logic, encoding dense computation on the hardware in the form of hardware/API-specific binaries like SPIR-V.

IREE Architecture

Workflow overviewlink

Using IREE involves the following general steps:

Import your model

Develop your program using one of the supported frameworks, then import into IREE
Select your deployment configuration

Identify your target platform, accelerator(s), and other constraints
Compile your model

Compile through IREE, picking settings based on your deployment configuration
Run your model

Use IREE's runtime components to execute your compiled model

Importing models from ML frameworkslink

IREE supports importing models from a growing list of ML frameworks and model formats:

Selecting deployment configurationslink

IREE provides a flexible set of tools for various deployment scenarios. Fully featured environments can use IREE for dynamic model deployments taking advantage of multi-threaded hardware, while embedded systems can bypass IREE's runtime entirely or interface with custom accelerators.

What platforms are you targeting? Desktop? Mobile? An embedded system?
What hardware should the bulk of your model run on? CPU? GPU?
How fixed is your model itself? Can the weights be changed? Do you want to support loading different model architectures dynamically?

IREE supports the full set of these configurations using the same underlying technology.

Compiling modelslink

Model compilation is performed ahead-of-time on a host machine for any combination of targets. The compilation process converts from layers and operators used by high level frameworks down into optimized native code and associated scheduling logic.

For example, compiling for GPU execution using Vulkan generates SPIR-V kernels and Vulkan API calls. For CPU execution, native code with static or dynamic linkage and the associated function calls are generated.

Running modelslink

IREE offers a low level C API, as well as several sets of API bindings for compiling and running programs using various languages.

Communication channelslink

GitHub issues: Feature requests, bugs, and other work tracking
IREE Discord server: Daily development discussions with the core team and collaborators
iree-discuss email list: Announcements, general and low-priority discussion

Roadmaplink

IREE is in the early stages of development and is not yet ready for broad adoption. We use both GitHub Projects and GitHub Milestones to track progress.

Pronounced "eerie" and often styled with the emoji ↩