Skip to main content

Ahead-of-time compilation toolkit

Project description

AOKit

Ahead-of-time compilation toolkit

aokit is a lightweight set of tools sitting on top of PyTorch ahead-of-time compilation APIs (e.g. torch.export.export and torch._inductor.aot*) to help organize, package, and distribute AoT artifacts within the Hugging Face ecosystem.

Extending PyTorch AOTInductor, aokit provides:

  • Regional compilation: compiles and loads individual submodules independently, with support for dynamic weight swapping even after quantization (e.g. with torchao)
  • Weight-less packaging: produces lightweight, composable, and reusable artifacts while decoupling graph and weights management
  • ZeroGPU compatibility: leverages lazy execution for seamless integration
  • kernels integration: automatically packages and loads required kernels (coming soon)

Installation

pip install aokit

Quickstart

Coming soon

API documentation

Coming soon

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

aokit-0.1.1.tar.gz (5.6 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

aokit-0.1.1-py3-none-any.whl (7.5 kB view details)

Uploaded Python 3

File details

Details for the file aokit-0.1.1.tar.gz.

File metadata

  • Download URL: aokit-0.1.1.tar.gz
  • Upload date:
  • Size: 5.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for aokit-0.1.1.tar.gz
Algorithm Hash digest
SHA256 93558a0f2e4ad16b478667eadf7b66ecd46bc43680249dafb1b54fe4d4a18742
MD5 57b85b694fe842eafdaeac138be56f8d
BLAKE2b-256 48e7a80f42302c1979a7f20db3f39247b11a572f49efb9eaad0fccd97c2e4df5

See more details on using hashes here.

Provenance

The following attestation bundles were made for aokit-0.1.1.tar.gz:

Publisher: release.yml on huggingface/aokit

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file aokit-0.1.1-py3-none-any.whl.

File metadata

  • Download URL: aokit-0.1.1-py3-none-any.whl
  • Upload date:
  • Size: 7.5 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for aokit-0.1.1-py3-none-any.whl
Algorithm Hash digest
SHA256 2bbd449bb5372e8c86d32b74e670c2b366f76000b9500e921c9b11f00958217f
MD5 297d0c7f96582a255f61e637a95624e5
BLAKE2b-256 f93c135172b6f2695418915ef74fb0fcd04006622c3c8c9b024d658b7d5ea5d3

See more details on using hashes here.

Provenance

The following attestation bundles were made for aokit-0.1.1-py3-none-any.whl:

Publisher: release.yml on huggingface/aokit

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page