# iree-org/iree

A retargetable MLIR-based machine learning compiler and runtime toolkit.

Repository: https://github.com/iree-org/iree
Canonical: https://ross.abutalabs.com/products/iree
Homepage: http://iree.dev/
Language: C++
License: Apache-2.0
License Family: permissive
Topics: mlir, vulkan, tensorflow, spirv, cuda, jax, pytorch, compiler, machine-learning, runtime, onnx, rocm
Last push: 2026-08-26T15:34:14+00:00

## Health v2 (maintenance only)
Score: 87/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 63, longevity 100
- inputs: {"age_days": 2547, "days_push": 7, "days_rel": 167, "gap_med": 40.5, "n_releases_24m": 15}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3901, forks 989 (observed 2026-08-28T04:08:28.544619+00:00)

## What it is
IREE is an MLIR-based end-to-end machine learning compiler and runtime that lowers models from frameworks like PyTorch, TensorFlow, JAX, and ONNX into a unified intermediate representation. It compiles ahead-of-time to efficient binaries for CPUs, GPUs, and accelerators, scaling from datacenter deployments down to mobile, edge, and bare-metal targets.

## Use cases
- compile pytorch models to run on gpu
- deploy machine learning models on mobile and edge devices
- run onnx models with low overhead runtime
- compile ml models for vulkan or cuda accelerators
- run ml inference on bare-metal embedded systems
- benchmark ml model inference across hardware targets
- convert tensorflow lite models to native binaries

## When to choose
- you need ahead-of-time compilation of ML models with small binary sizes (down to ~30KB) for embedded or edge targets
- you need to deploy models across diverse hardware including CUDA, ROCm/HIP, Vulkan, and Metal from a single toolchain
- you need dynamic shapes, streaming, and control flow support in compiled models
- you want profiling, debugging, and bare-metal deployment support for ML workloads

## When to avoid
- you just need a simple inference runtime for one framework on one platform - framework-native runtimes are simpler
- you need training support - IREE focuses on inference deployment, not training
- you want a turnkey solution without understanding compilers or MLIR concepts
- your target platform is not in the support matrix and you cannot contribute support

## Facets
- artifact type: library
- maturity: active
- function: compiler, machine-learning, llm-inference, gpu-computing, interpreter, benchmarking
- domain: machine-learning, compilers, deep-learning, gpu-computing, embedded-systems, developer-tools
- platform: windows, cross-platform, wasm, cpp, python
- tags: mlir, ml-model-deployment, aot-compilation, vulkan, cuda, rocm, metal, spirv, edge-deployment, hardware-abstraction-layer, linux, macos, android, ios, gpu

## Member repositories
- iree-org/iree (main) score 87

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:28.544619+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:25:10.501672+00:00, confidence not recorded.
  - readme: https://github.com/iree-org/iree (fetched 2026-08-28T04:08:28.544619+00:00, sha 649a4ee800a5)
  - homepage: http://iree.dev/ (fetched 2026-08-29T09:19:24.633752+00:00, sha f226701274f7)
  - site_page: https://iree.dev/building-from-source/getting-started (fetched 2026-08-29T09:19:24.636713+00:00, sha 5aa133185cc1)
  - site_page: https://iree.dev/developers/design-docs/design-roadmap (fetched 2026-08-29T09:19:24.640822+00:00, sha cbc9e761fe49)
  - site_page: https://iree.dev/developers/design-docs/function-abi (fetched 2026-08-29T09:19:24.643857+00:00, sha 636389d81b08)
  - site_page: https://iree.dev/developers/design-docs/invocation-execution-model (fetched 2026-08-29T09:19:24.645623+00:00, sha a37549f42fdc)
  - site_page: https://iree.dev/developers/design-docs/cuda-hal-driver (fetched 2026-08-29T09:19:24.647751+00:00, sha d855208bc5df)
  - site_page: https://iree.dev/developers/design-docs/hip-hal-driver (fetched 2026-08-29T09:19:24.649643+00:00, sha 6e26ce3e837d)
  - site_page: https://iree.dev/developers/design-docs/metal-hal-driver (fetched 2026-08-29T09:19:24.651618+00:00, sha 678935ef570d)
  - site_page: https://iree.dev/developers/debugging/releases (fetched 2026-08-29T09:19:24.638843+00:00, sha c1f91b718496)
- Data as of 2026-08-30T08:39:29.467469+00:00.