Skip to main content

The latest ZenDNN Plugin for PyTorch* (zentorch) 2.13.0.3 is here!

The ZenDNN PyTorch plugin is called zentorch. Combined with PyTorch's torch.compile, zentorch transforms deep learning pipelines into finely-tuned, AMD-specific engines, delivering unparalleled efficiency and speed for large-scale inference workloads.

This upgrade continues the focus on optimizing inference with Recommender Systems and Large Language Models on AMD EPYC™ CPUs. It includes AMD EPYC™ enhancements for bfloat16 performance, expanded support for cutting-edge models like Llama 3.2 and 3.3, Microsoft Phi, and more as well as support for a wide-variety of quantization configurations. The quantization support included 4-bit weight-only quantization, along with support for INT8 dynamic activation and INT8 weight quantization, and quantized support for the DLRM-v2 model with a mix of 8-bit and 4-bit quantization. This also includes support for running generative models with vLLM. This release introduces functional support for running LLMs using float16 precision with vLLM on 6th Gen AMD EPYC™ processors.

Under the hood, ZenDNN’s enhanced AMD-specific optimizations operate at every level. In addition to highly optimized operator microkernels, these include comprehensive graph optimizations including pattern identification, graph reordering, and fusions. They also incorporate optimized embedding bag kernels and enhanced zenMatMul matrix splitting strategies which leverage the AMD EPYC™ microarchitecture to deliver enhanced throughput and latency.

The vLLM-ZenTorch plugin extends these benefits to the vLLM inference engine, enabling plug-and-play acceleration of large language model inference on AMD EPYC™ CPUs. By integrating ZenTorch with vLLM, users can achieve significant throughput improvements for LLM workloads with zero code changes.

The zentorch plugin 2.13.0.3 seamlessly works with PyTorch 2.13.0+cpu, offering a high-performance experience for deep learning on AMD EPYC™ platforms.

In addition to stable (GA) releases, the zentorch plugin provides weekly releases via the zentorch-weekly package on PyPI.

Support

We welcome feedback, suggestions, and bug reports. Should you have any of the these, please kindly file an issue on the ZenDNN Plugin for PyTorch GitHub page here

License

AMD copyrighted code in ZenDNN is subject to the Apache-2.0, MIT, or BSD-3-Clause licenses; consult the source code file headers for the applicable license. Third party copyrighted code in ZenDNN is subject to the licenses set forth in the source code file headers of such code.

Build Information

Field Value
Source Tag 2026-10-14-22-07
Tag Commit aea71fa085b0cc211a36f7fd5d594fe27cfb8a73
Build Commit aea71fa085b0cc211a36f7fd5d594fe27cfb8a73
PyTorch Version 2.13.0+cpu
Release Type ga

Built from 2026-10-14-22-07

Metadata

Release files for zentorch 2.13.0.3

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Built distributions (wheels)

Table of built distributions (wheels) for zentorch 2.13.0.3
File Actions
zentorch-2.13.0.3-cp314-cp314-manylinux_2_28_x86_64.whl CPython 3.14 CPython 3.14 Linux glibc 2.28+ x86-64 Details for zentorch-2.13.0.3-cp314-cp314-manylinux_2_28_x86_64.whl
zentorch-2.13.0.3-cp313-cp313-manylinux_2_28_x86_64.whl CPython 3.13 CPython 3.13 Linux glibc 2.28+ x86-64 Details for zentorch-2.13.0.3-cp313-cp313-manylinux_2_28_x86_64.whl
zentorch-2.13.0.3-cp312-cp312-manylinux_2_28_x86_64.whl CPython 3.12 CPython 3.12 Linux glibc 2.28+ x86-64 Details for zentorch-2.13.0.3-cp312-cp312-manylinux_2_28_x86_64.whl
zentorch-2.13.0.3-cp311-cp311-manylinux_2_28_x86_64.whl CPython 3.11 CPython 3.11 Linux glibc 2.28+ x86-64 Details for zentorch-2.13.0.3-cp311-cp311-manylinux_2_28_x86_64.whl
zentorch-2.13.0.3-cp310-cp310-manylinux_2_28_x86_64.whl CPython 3.10 CPython 3.10 Linux glibc 2.28+ x86-64 Details for zentorch-2.13.0.3-cp310-cp310-manylinux_2_28_x86_64.whl

Total release size: 119.3 MB

Release files / zentorch-2.13.0.3-cp314-cp314-manylinux_2_28_x86_64.whl

Download URL zentorch-2.13.0.3-cp314-cp314-manylinux_2_28_x86_64.whl
Size 23.9 MB
Tags CPython 3.14 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
e628e4758726cd542f48e2731edc25cc33dfd8f4b14279916a88c1ea4e10244f
BLAKE2b-256 checksum
How to use checksums
b2c1f17cf34b46c9417df62075e023a1813011571a516cc124b023d56909eb6c
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.10.20

Release files / zentorch-2.13.0.3-cp313-cp313-manylinux_2_28_x86_64.whl

Download URL zentorch-2.13.0.3-cp313-cp313-manylinux_2_28_x86_64.whl
Size 23.9 MB
Tags CPython 3.13 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
bee5b93d9b3e4cb8217e159a5e92842646fc0e5b9452f6a43e1aa88ba214b292
BLAKE2b-256 checksum
How to use checksums
972a3f4c27b2d8a4557a401fbf5ef67c995d9b6b8c342fa060acb6410024f18d
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.10.20

Release files / zentorch-2.13.0.3-cp312-cp312-manylinux_2_28_x86_64.whl

Download URL zentorch-2.13.0.3-cp312-cp312-manylinux_2_28_x86_64.whl
Size 23.9 MB
Tags CPython 3.12 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
86d3c888c33bef94efcbf84e3205d2420061052d0344907b0fa960e767b1ab73
BLAKE2b-256 checksum
How to use checksums
d4f01a17bb95e3576fd4082aa6dfb6d70279e6f0e0b060e6aed15525152a5054
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.10.20

Release files / zentorch-2.13.0.3-cp311-cp311-manylinux_2_28_x86_64.whl

Download URL zentorch-2.13.0.3-cp311-cp311-manylinux_2_28_x86_64.whl
Size 23.9 MB
Tags CPython 3.11 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
18c9519cae3f4f43b680977a0b46ed522487d52cf507c997aaa216035b0172bf
BLAKE2b-256 checksum
How to use checksums
b6d3910f74082340651adaded176087458e769fe66f9988a29db5b5587584ca6
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.10.20

Release files / zentorch-2.13.0.3-cp310-cp310-manylinux_2_28_x86_64.whl

Download URL zentorch-2.13.0.3-cp310-cp310-manylinux_2_28_x86_64.whl
Size 23.9 MB
Tags CPython 3.10 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
7ed8638d136702403f4ec36c80b6d3484109b7cb103c8c8fbf3f74f3ddbce650
BLAKE2b-256 checksum
How to use checksums
40a7e94f9ba0ecc1c50b022f4c589d32a1ba7daef66b680c16b5443bb96a713d
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.10.20

Security for zentorch 2.13.0.3

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

2.13.0.3 This release

5 release files No provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page