# microsoft/DirectML

⚠️DirectML is in maintenance mode ⚠️ DirectML is a high-performance, hardware-accelerated DirectX 12 library for machine learning. DirectML provides GPU acceleration for common machine learning tasks across a broad range of supported hardware and drivers, including all DirectX 12-capable GPUs from vendors such as AMD, Intel, NVIDIA, and Qualcomm.

Repository: https://github.com/microsoft/DirectML
Canonical: https://ross.abutalabs.com/products/directml
Language: C++
License: MIT
License Family: permissive
Last push: 2026-04-27T23:43:38+00:00

## Health v2 (maintenance only)
Score: 58/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 79, release rhythm 8, longevity 100
- inputs: {"age_days": 2732, "days_push": 128, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2580, forks 339 (observed 2026-08-28T04:07:02.389898+00:00)

## What it is
DirectML is a high-performance, hardware-accelerated DirectX 12 library for machine learning that provides GPU acceleration across DirectX 12-capable GPUs from AMD, Intel, NVIDIA, and Qualcomm. It is now in maintenance mode, with Microsoft recommending Windows ML for Windows 11 24H2 and later.

## Use cases
- run machine learning models with GPU acceleration on any DirectX 12 GPU
- accelerate inference in games and real-time applications on Windows
- execute ONNX models on Windows GPUs without vendor-specific SDKs
- use PyTorch or TensorFlow with a DirectX 12 backend
- build low-latency ML features into Windows desktop applications
- run ML workloads across AMD, Intel, NVIDIA, and Qualcomm GPUs with one API

## When to choose
- you need GPU-accelerated ML on Windows across heterogeneous GPU vendors
- you target older Windows 10/11 releases where DirectML ships as an OS component
- you want low-overhead DirectX 12 interoperability in games or real-time apps

## When to avoid
- you target Windows 11 24H2 or later and can use Windows ML instead
- you need new features or active development, since DirectML is in maintenance mode
- you develop primarily on Linux or macOS

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, gpu-computing, llm-inference
- domain: machine-learning, gpu-computing, windows
- platform: windows, cpp
- tags: directx-12, gpu-acceleration, onnx-runtime, pytorch, tensorflow, microsoft, gpu

## Member repositories
- microsoft/DirectML (main) score 58

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:02.389898+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 2026-08-30T02:22:39.057775+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/DirectML (fetched 2026-08-28T04:07:02.389898+00:00, sha 7f32b2120a50)
- Data as of 2026-08-30T08:39:29.467469+00:00.