ZenDNN library, which includes APIs for basic neural network building blocks optimized for AMD CPU architecture, enables deep learning application and framework developers to improve deep learning inference performance on AMD CPUs.
For the current version, refer to the ZenDNN page.
Support
For support options, refer to Technical Support.
- 5.1
- 5.0.2
- 5.0.1
- 5.0
- 4.2
- 4.1
- 4.0
- 3.3
5.1 Release Highlights
- Framework Compatibility
- PyTorch & TensorFlow: We've added full compatibility with PyTorch 2.7 and TensorFlow 2.19, ensuring seamless integration with the latest versions of these leading AI frameworks.
- vLLM + zentorch Plugin: The new zentorch plugin for vLLM delivers a significant performance uplift of up to 21% on a variety of models compared to vLLM-IPEX.
- Java® Integration: We've enabled support for PluggableDevice in TensorFlow-Java, a feature essential for zentf functionality. This feature has been officially contributed and upstreamed to the TensorFlow-Java repository, strengthening its core capabilities. For more details, please see the TensorFlow-Java integration Blog.
- Performance Optimizations
- Recommender Systems: We've introduced several key optimizations to boost the performance of recommender models, such as DLRMv2.
- EmbeddingBag Improvements: EmbeddingBag Improvements: New "out" variants of EmbeddingBag and related operators now write directly to a shared output buffer, eliminating the need for a separate concatenation operation and improving efficiency.
- Concat Optimization: We've introduced a new optimization that fuses the concatenation operation after Bottom MLP and EmbeddingBag, for the DLRMv2 model.
- New Operator Fusions: We've added new operator fusions to accelerate common computational patterns, resulting in a 25% performance uplift for the DIEN BF16 model.
- MatMul + BiasAdd + Tanh
- MatMul + BiasAdd + Sigmoid
- Kernel Optimizations:
- BF16/FP32 MatMul: A new kernel for BF16/FP32 matrix multiplication has been introduced that eliminates overheads in less compute-intensive GEMM operations, leading to improved performance of the DIEN model.
- Ahead of Time (AOT) Reorder: We now support AOT reordering for MatMul kernels across INT8, BF16, and FP32 data types.
- ZenDNN Enhancements: Added support for MatMul(+fused) Low Overhead API (LOA) to improve performance of small matrix shapes, further improving performance and efficiency.
- Ecosystem Contribution
- We are actively contributing our optimization work directly to the core PyTorch codebase, as well as the PluggableDevice feature to the TensorFlow-Java repository. These regular upstream contributions strengthen the native performance and capabilities of both frameworks, benefiting the entire community.
Resources
Version |
Resources |
| 5.1 | ZenDNN 5.1 User Guide ZenDNN 5.1 Support Matrix |
Archives v5.1
| ZenDNN Plug-in for PyTorch | Description | MD5SUM |
| ZENTORCH_v5.1.0_Python_v3.10.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.10 | da8bf1d3d4f5975ef17d8bffab790f55 |
| ZENTORCH_v5.1.0_Python_v3.11.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.11 | 608699f120980469bd2fc8c5cfb6395f |
| ZENTORCH_v5.1.0_Python_v3.12.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.12 | d119f6e6551083663486367170ebca4d |
| ZENTORCH_v5.1.0_Python_v3.13.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.13 | f8abf43aff5af4790a2977972fbffca0 |
| ZENTORCH_v5.1.0_Python_v3.9.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.9 | 0f858c78272de09b95719c76ab68d1b4 |
| ZenDNN Plug-in for TensorFlow | Description | MD5SUM |
| ZENTF_v5.1.0_Python_v3.10.zip | This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.10 | 885c7df96ec9b3f2b3302ab3f7bfdef1 |
| ZENTF_v5.1.0_Python_v3.11.zip | This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.11 | 2a77a62d110ae96244f1dc34ddcbda64 |
| ZENTF_v5.1.0_Python_v3.12.zip | This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.12 | 81799cd80331155555105efdf10f10aa |
| ZENTF_v5.1.0_Python_v3.9.zip | This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.9 | 745fbefd95f761a0fd0cac3bd2339289 |
| ZENTF_v5.1.0_C++_API.zip | This zip file contains the ZenDNN TensorFlow Plug-in with C++ APIs | 1824f52455c76888fca433028f943414 |
5.0.2 Release Highlights
- Framework Compatibility: Fully compatible with PyTorch 2.6 and TensorFlow 2.18.
- Java® Integration: Introduces a Java interface to the TensorFlow plugin (zentf) via TensorFlow Java.
- Optimized Quantized Model Support: Enhanced performance for INT8/INT4-quantized DLRM models.
Resources
| Version | Resources |
| 5.0.2 |
Archives v5.0.2
| Title | Description | MD5SUM |
| ZenDNN Plug-in for PyTorch | ||
| ZENTORCH_v5.0.2_Python_v3.10.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.10 | ad694dae86c827c5069f370a76680c4e |
| ZENTORCH_v5.0.2_Python_v3.11.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.11 | 3a768db8d8e72e14260530b865d4affd |
| ZENTORCH_v5.0.2_Python_v3.12.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.12 | dc4af7dfa8a3223f8f1939e1727c9daa |
| ZENTORCH_v5.0.2_Python_v3.13.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.13 | 5f7cca5a9e8161321bac81c5a4061672 |
| ZENTORCH_v5.0.2_Python_v3.9.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.9 | e67593eeb51e22d3451839aef9b3f9c2 |
| ZenDNN Plug-in for TensorFlow |
||
| ZENTF_v5.0.2_Python_v3.10.zip | This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.10 | a5668caa29fffcd4d8dfabb845228e7b |
| ZENTF_v5.0.2_Python_v3.11.zip | This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.11 | 6fb4765983a4c9f27d18cd4d670a3d2d |
| ZENTF_v5.0.2_Python_v3.12.zip | This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.12 | bbbd9bcccccd87b4f65b515f4076993e |
| ZENTF_v5.0.2_Python_v3.9.zip | This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.9 | f99467d7bfd034d6ec4ccb3e2c9998f7 |
| ZENTF_v5.0.2_C++_API.zip | This zip file contains the ZenDNN TensorFlow Plug-in with C++ APIs | 387ce4f857936132d4ca410d66270836 |
5.0.1 Release Highlights
- Compatible with deep-learning frameworks: Aligned closely with PyTorch 2.5 and TensorFlow 2.18, helping ensure smooth upgrades and interoperability.
- Efficient Model Execution: Added support for INT8/INT4-quantized DLRM models in zentorch, unlocking faster inference with lower memory usage compared to BF16-precision. This release supports the MLPerf® version of DLRMv2; support for generic models are planned for the next release.
Resources
| Version | Resources |
| 5.0.1 | ZenDNN 5.0.1 User Guide Support Matrix |
Archives v5.0.1
| Title | Description | MD5SUM |
| ZenDNN Plug-in for PyTorch | ||
| ZENTORCH_v5.0.1_Python_v3.10.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.10 | 8a0fb79874dcfe4537db8e7b898fb3ad |
| ZENTORCH_v5.0.1_Python_v3.11.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.11 | 3452c37545cd7d1748e01d3ba2022810 |
| ZENTORCH_v5.0.1_Python_v3.12.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.12 | 988260afb5ed25d0f2f11b22d08d11f9 |
| ZENTORCH_v5.0.1_Python_v3.9.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.9 | ed674d7526ab244fa31fad57f3b34413 |
| ZenDNN Plug-in for TensorFlow |
||
| ZENTF_v5.0.1_Python_v3.10.zip | This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.10 | 227b1bd6680493e683ef51bbc09df007 |
| ZENTF_v5.0.1_Python_v3.11.zip | This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.11 | 792b90b68bdc9eb2bbc15d5158bb724c |
| ZENTF_v5.0.1_Python_v3.12.zip | This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.12 | 75016fa0d2f05681eac047ec2931a629 |
| ZENTF_v5.0.1_Python_v3.9.zip | This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.9 | c134e7da17b6a6e7c89370440a8905ba |
| ZENTF_v5.0.1_C++_API.zip | This zip file contains the ZenDNN TensorFlow Plug-in with C++ APIs | 56687fb93f3fd87e742b49a621b7d050 |
| MLPerf DLRM V2 Benchmark Binaries | ||
| dlrmv2-mlperf-binary-updated.zip | This zip file contains the zentorch wheel file and the necessary scripts to run MLPerf DLRM v2 benchmarking on AMD EPYC. The binary is intended solely for MLPerf DLRM v2 benchmarking purposes on AMD EPYC CPU. For optimized inference performance on AMD EPYC CPUs, please use the ZenDNN software stack. | a94a33a7be43e994def63d3ad5a3d5b1 |
5.0 Release Highlights
- Support for 5th Gen AMD EPYC™ processors, formerly codenamed “Turin”
- Framework Support: PyTorch 2.4.0, TensorFlow 2.17 and ONNXRT 1.19.2
- New APIs in the ZenDNN Plugin for PyTorch (zentorch), such as zentorch.llm.optimize() and zentorch.load_woq_model(), for enhanced LLM performance
- Enhanced matmul operators and fusions and a new BF16 auto-tuning algorithm targeted for generative LLMs.
- An optimized Scalar Dot Product Attention operator including-KV cache performance optimizations tailored to AMD EPYC™ cache architectures
- Support for INT4 Weight-Only-Quantization (WOQ)
- Improved Model Support: Llama3.1 and 3.2, Phi3, ChatGLM3, Qwen2, GPT-J
Release Blog
ZenDNN 5.0: Supercharge AI on AMD EPYC™ Server CPUs
Resources
| Version | Resources |
| 5.0 | ZenDNN 5.0 User Guide Support Matrix |
Archives v5.0
| Title | Description | MD5SUM |
| ZenDNN Plug-in for PyTorch | ||
| ZENTORCH_v5.0.0_Python_v3.10.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.10 | f34e7b450fe0d6bb973db41838430bf0 |
| ZENTORCH_v5.0.0_Python_v3.11.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.11 | a49446852ab4b529e70b234b95a8c47f |
| ZENTORCH_v5.0.0_Python_v3.8.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.8 | c8c2deca8fefe6c7d6699bda20dd95a3 |
| ZENTORCH_v5.0.0_Python_v3.9.zip | This zip file contains the zentorch wheel file and the necessary scripts to set up the environment variables. Compatible with Python version 3.9 | ffdfa8a4b65cbf0547790fc60350761c |
| ZenDNN Plug-in for TensorFlow |
||
| ZENTF_v5.0.0_Python_v3.10.zip | This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.10 | f5624742f5e5a85aa2b43cc1706d1a6c |
| ZENTF_v5.0.0_Python_v3.11.zip | This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.11 | 043fccc676b20d10d7f7ba4719dd89cd |
| ZENTF_v5.0.0_Python_v3.12.zip | This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.12 | 6602d4ce1617bc05f0169d0c2f93e4ee |
| ZENTF_v5.0.0_Python_v3.9.zip | This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.9 | 4a33b8e27f4544f9e36421c56a6fdea3 |
| ZENTF_v5.0.0_C++_API.zip | This zip file contains the ZenDNN TensorFlow Plug-in with C++ APIs | fdb264a921a81f56965f4294d23e7abf |
| ZenDNN with ONNX Runtime | ||
| ONNXRT_v1.19.2_ZenDNN_v5.0_Python_v3.10.zip | This zip file contains the ONNX Runtime wheel and the necessary scripts to set up the environment variables. Compatible with Python 3.10 | 38f64749c2e3e6d249b0c38bad2b0400 |
| ONNXRT_v1.19.2_ZenDNN_v5.0_Python_v3.11.zip | This zip file contains the ONNX Runtime wheel and the necessary scripts to set up the environment variables. Compatible with Python 3.11 | 6368acd895504df095017d25dcba774e |
| ONNXRT_v1.19.2_ZenDNN_v5.0_Python_v3.8.zip | This zip file contains the ONNX Runtime wheel and the necessary scripts to set up the environment variables. Compatible with Python 3.8 | 2b2d0167497025c059bd8572e3c2ebd6 |
| ONNXRT_v1.19.2_ZenDNN_v5.0_Python_v3.9.zip | This zip file contains the ONNX Runtime wheel and the necessary scripts to set up the environment variables. Compatible with Python 3.9 | c996d54ecda4783ea5e78acb51026497 |
| ONNXRT_v1.19.2_ZenDNN_v5.0_C++_API.zip | This zip file contains the ZenDNN ONNX Runtime with C++ APIs | a3cb5654e7877db81abb1189b8bc3074 |
4.2 Release Highlights
- Plugins to native PyTorch and TensorFlow
- Framework Support for PyTorch 2.1, TensorFlow 2.16 and ONNXRT 1.17.0
- Enhancements with microkernels, mempool optimizations, and efficient multi-threading on the large number of AMD EPYC™ cores
- New INT8 support
- Improved algorithms for embedding bags
- Improved GEMM algorithms
Release Blog
ZenDNN 4.2: Introducing a New Plugin Architecture
Resources
| Version | Resources |
| 4.2 | ZenDNN 4.2 User Guide Support Matrix |
Archives v4.2
| Title | Description | MD5SUM |
| ZenDNN Plug-in for PyTorch | ||
| ZENTORCH_v4.2.0_Python_v3.10.zip | This zip file contains the zentorch wheel file compatible with Python version 3.10 | 36464127bb6d7d691d06662cdff667d7 |
| ZENTORCH_v4.2.0_Python_v3.11.zip |
This zip file contains the zentorch wheel file compatible with Python version 3.11 |
7873b9a54ccb258c985e79f1e59e4921 |
| ZENTORCH_v4.2.0_Python_v3.8.zip |
This zip file contains the zentorch wheel file compatible with Python version 3.8 |
2c26985eb8d30d1daaf64380c5a7c92d |
| ZENTORCH_v4.2.0_Python_v3.9.zip |
This zip file contains the zentorch wheel file compatible with Python version 3.9 |
39cad1529e3d92109dfd29dde5542412 |
| ZenDNN Plug-in for TensorFlow |
||
| ZENTF_v4.2.0_Python_v3.10.zip |
This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.10 |
7819c979e1147861395634c29da2dbb0 |
| ZENTF_v4.2.0_Python_v3.11.zip |
This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.11 |
b34c6c7fda3579cdeb76756db23146fd |
| ZENTF_v4.2.0_Python_v3.12.zip |
This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.12 |
5a7aacda4f99dae485920e1d7cdf307b |
| ZENTF_v4.2.0_Python_v3.9.zip |
This zip file contains the zentf wheel file and the necessary scripts to set up the environment variables. Compatible with Python 3.9 |
365ef04189b615ecfc0a2e8c1481f013 |
| ZENTF_v4.2.0_C++_API.zip | This zip file contains the ZenDNN Plug-in with C++ APIs | 4802d21e31e2e8f9ca632e987889f7d4 |
| ZenDNN with ONNX Runtime | ||
| ONNXRT_v1.17.0_ZenDNN_v4.2_C++_API.zip | This zip file contains the ONNX Runtime with C++ APIs | 4b12aa577e8919b6e08a0758a8cc0fa1 |
| ONNXRT_v1.17.0_ZenDNN_v4.2_Python_v3.10.zip | This zip file contains the ONNX Runtime wheel and the necessary scripts to set up the environment variables. Compatible with Python 3.10 | ace337e60f130cb7d8e0a1fc6840848b |
| ONNXRT_v1.17.0_ZenDNN_v4.2_Python_v3.11.zip | This zip file contains the ONNX Runtime wheel and the necessary scripts to set up the environment variables. Compatible with Python 3.11 | fd6540178fa87c37048239107b4f4f4d |
| ONNXRT_v1.17.0_ZenDNN_v4.2_Python_v3.8.zip | This zip file contains the ONNX Runtime wheel and the necessary scripts to set up the environment variables. Compatible with Python 3.8 | 0a6b531c1ec709207c82dec7e4d0b98a |
| ONNXRT_v1.17.0_ZenDNN_v4.2_Python_v3.9.zip | This zip file contains the ONNX Runtime wheel and the necessary scripts to set up the environment variables. Compatible with Python 3.9 | d00350e76d4fc08486b5213e4c2181e4 |
Resources
| Version | Resources |
| 4.1 | ZenDNN 4.1 User Guide Support Matrix |
Archives v4.1
TensorFlow, PyTorch, and ONNX Runtime are available with different Python versions.
| File Name | Version | Size (MB) | Launch Date | OS | Bitness | Description |
|---|---|---|---|---|---|---|
| TensorFlow | ||||||
| TF_v2.12_ZenDNN_v4.1_Python_v3.8.zip | 4.1 | 251 | 9/25/2023 | Ubuntu/RHEL | 64 | TensorFlow v2.12 + ZenDNN binary release package with python v3.8. MD5 Checksum: 460232df133cacda36ec5474cce09502 |
| TF_v2.12_ZenDNN_v4.1_Python_v3.9.zip | 4.1 | 251 | 9/25/2023 | Ubuntu/RHEL | 64 | TensorFlow v2.12 + ZenDNN binary release package with python v3.9. MD5 Checksum: 92c74a43eb64854ad8398ca88f22b151 |
| TF_v2.12_ZenDNN_v4.1_Python_v3.10.zip | 4.1 | 251 | 9/25/2023 | Ubuntu/RHEL | 64 | TensorFlow v2.12 + ZenDNN binary release package with python v3.10. MD5 Checksum: 52623e6708043f1fca9664c259757e03 |
| TF_v2.12_ZenDNN_v4.1_Python_v3.11.zip | 4.1 | 251 | 9/25/2023 | Ubuntu/RHEL | 64 | TensorFlow v2.12 + ZenDNN binary release package with python v3.11. MD5 Checksum: 1c5afea80d5c5165c10069d413fb55b6 |
| TF_v2.12_ZenDNN_v4.1_C++_API.zip | 4.1 | 199 | 9/25/2023 | Ubuntu/RHEL | 64 | TensorFlow v2.12 + ZenDNN binary release package for C++ API interface. MD5 Checksum: 3219d8d137c60d306c20668179bb188e |
| PyTorch | ||||||
| PT_v1.13_ZenDNN_v4.1_Python_v3.7.zip | 4.1 | 115 | 9/25/2023 | Ubuntu/RHEL | 64 | PyTorch v1.13 + ZenDNN binary release package with python v3.7. MD5 Checksum: 17a07cc5658ca90e094cb632af6cbf8a |
| PT_v1.13_ZenDNN_v4.1_Python_v3.8.zip | 4.1 | 115 | 9/25/2023 | Ubuntu/RHEL | 64 | PyTorch v1.13+ ZenDNN binary release package with python v3.8. MD5 Checksum: 0e386b480b3448ace8d09c312775a5af |
| PT_v1.13_ZenDNN_v4.1_Python_v3.9.zip | 4.1 | 114 | 9/25/2023 | Ubuntu/RHEL | 64 | PyTorch v1.13 + ZenDNN binary release package with python v3.9. MD5 Checksum: 89b479a65d2f6d51d517ea6322b6cd2f |
| PT_v1.13_ZenDNN_v4.1_Python_v3.10.zip | 4.1 | 114 | 9/25/2023 | Ubuntu/RHEL | 64 | PyTorch v1.13 + ZenDNN binary release package with python v3.10. MD5 Checksum: 0d817429f6b233e80d58af9bedb66708 |
| PT_v1.13_ZenDNN_v4.1_C++_API.zip | 4.1 | 74 | 9/25/2023 | Ubuntu/RHEL | 64 | PyTorch v1.13 + ZenDNN binary release package for C++ API interface. MD5 Checksum: 0584e8823ddbe2d76b0a507675715db5 |
| ONNX Runtime | ||||||
| ONNXRT_v1.15.1_ZenDNN_v4.1_Python_v3.8.zip | 4.1 | 21 | 9/25/2023 | Ubuntu/RHEL | 64 | ONNX Runtime v1.15.1 + ZenDNN binary release package with python v3.8. MD5 Checksum: 3f521d8174484d811c615e1657e8231a |
| ONNXRT_v1.15.1_ZenDNN_v4.1_Python_v3.9.zip | 4.1 | 21 | 9/25/2023 | Ubuntu/RHEL | 64 | ONNX Runtime v1.15.1 + ZenDNN binary release package with python v3.9. MD5 Checksum: 541fdf9ed5292ff15eafbec75520bb76 |
| ONNXRT_v1.15.1_ZenDNN_v4.1_Python_v3.10.zip | 4.1 | 21 | 9/25/2023 | Ubuntu/RHEL | 64 | ONNX Runtime v1.15.1 + ZenDNN binary release package with python v3.10. MD5 Checksum: 9ed490d93f736bc251704f8a49f98c99 |
| ONNXRT_v1.15.1_ZenDNN_v4.1_Python_v3.11.zip | 4.1 | 21 | 9/25/2023 | Ubuntu/RHEL | 64 | ONNX Runtime v1.15.1 + ZenDNN binary release package with python v3.11. MD5 Checksum: 8b23cbb7f659a9806e8965495811862e |
| ONNXRT_v1.15.1_ZenDNN_v4.1_C++_API.zip | 4.1 | 30 | 9/25/2023 | Ubuntu/RHEL | 64 | ONNX Runtime v1.15.1 + ZenDNN binary release package for C++ API interface. MD5 Checksum: d977b9a223e128689741650e9709a0ed |
| ONNX Runtime Windows | ||||||
| ONNXRT_v1.15.1_ZenDNN_v4.1_Python_v3.8_Win.zip | 4.1 | 13 | 9/25/2023 | Windows | 64 | ONNX Runtime v1.15.1 + ZenDNN binary release package with python v3.8 for Windows OS. MD5 Checksum: 8cd35dc54c94b52725a0cd7c30796c9c |
| ONNXRT_v1.15.1_ZenDNN_v4.1_Python_v3.9_Win.zip | 4.1 | 13 | 9/25/2023 | Windows | 64 | ONNX Runtime v1.15.1 + ZenDNN binary release package with python v3.9 for Windows OS. MD5 Checksum: 58c13fe5851b85ab07ff9068dfb49955 |
| ONNXRT_v1.15.1_ZenDNN_v4.1_Python_v3.10_Win.zip | 4.1 | 13 | 9/25/2023 | Windows | 64 | ONNX Runtime v1.15.1 + ZenDNN binary release package with python v3.10 for Windows OS. MD5 Checksum: d3985ea8ffa5fdce938cb793b8b46ce4 |
| ONNXRT_v1.15.1_ZenDNN_v4.1_Python_v3.11_Win.zip | 4.1 | 13 | 9/25/2023 | Windows | 64 | ONNX Runtime v1.15.1 + ZenDNN binary release package with python v3.11 for Windows OS. MD5 Checksum: f64585efdc6144a1ab9ad7405f9cd1e6 |
| ONNXRT_v1.15.1_ZenDNN_v4.1_C++_API_Win.zip | 4.1 | 12 | 9/25/2023 | Windows | 64 | ONNX Runtime v1.15.1 + ZenDNN binary release package for C++ API interface for Windows OS. MD5 Checksum: 59f69a0aba8a86bbf3bb542163d13866 |
Resources
| Version | Resources |
| 4.0 | ZenDNN 4.0 – Documents |
Archives v4.0
| File Name | Version | Size | Launch Date | OS | Bitness | Description |
| TensorFlow | ||||||
| TF_v2.10_ZenDNN_v4.0_Python_v3.7.zip | 4 | 253 | 1/20/2023 | Ubuntu/ RHEL | 64 | TensorFlow v2.10 + ZenDNN binary release package with python v3.7. MD5 Checksum: 6dd12b66a58c2fcf0b72dd9b6334b4fe |
| TF_v2.10_ZenDNN_v4.0_Python_v3.8.zip | 4 | 253 | 1/20/2023 | Ubuntu/RHEL | 64 | TensorFlow v2.10 + ZenDNN binary release package with python v3.8. MD5 Checksum: a5f2f72bf92c7c6a96bb79dcca8e6501 |
| TF_v2.10_ZenDNN_v4.0_Python_v3.9.zip | 4 | 253 | 1/20/2023 | Ubuntu/RHEL | 64 | TensorFlow v2.10 + ZenDNN binary release package with python v3.9. MD5 Checksum: 85a5e173143087189be1e11118324c40 |
| TF_v2.10_ZenDNN_v4.0_Python_v3.10.zip | 4 | 253 | 1/20/2023 | Ubuntu/RHEL | 64 | TensorFlow v2.10 + ZenDNN binary release package with python v3.10. MD5 Checksum: d0b17b669fe9385bc643d988fd075598 |
| TF_v2.10_ZenDNN_v4.0_C++_API.zip | 4 | 474 | 1/20/2023 | Ubuntu/RHEL | 64 | TensorFlow v2.10 F v2.10 + ZenDNN binary release package for C++ API interface. MD5 Checksum: bb05ff87083877bc00e14b42c16b8fe2 |
| TF_v2.10_ZenDNN_v4.0_source_code.tar.gz | 4 | 130 | 1/20/2023 | Ubuntu/RHEL | 64 | Integration source code of TensorFlow v2.10 with ZenDNN. The instruction file inside zip code describe how to use the integration source code and how to build TensorFlow v2.10 with ZenDNN. MD5 Checksum:bd9b704b9b3338ba98fa59047c23d373 |
| PyTorch | ||||||
| PT_v1.12_ZenDNN_v4.0_Python_v3.7.zip | 4 | 112 | 1/20/2023 | Ubuntu/RHEL | 64 | PyTorch v1.12 + ZenDNN binary release package with python v3.7. MD5 Checksum: 7659431337d5b15c2fe97d6a7407bc4f |
| PT_v1.12_ZenDNN_v4.0_Python_v3.8.zip | 4 | 112 | 1/20/2023 | Ubuntu/RHEL | 64 | PyTorch v1.12+ ZenDNN binary release package with python v3.8. MD5 Checksum: ce4fc1d5d780d2f279ad629fa11f32ff |
| PT_v1.12_ZenDNN_v4.0_Python_v3.9.zip | 4 | 112 | 1/20/2023 | Ubuntu/RHEL | 64 | PyTorch v1.12 + ZenDNN binary release package with python v3.9. MD5 Checksum: 0b6a45fd63483355714793794f1db990 |
| PT_v1.12_ZenDNN_v4.0_Python_v3.10.zip | 4 | 112 | 1/20/2023 | Ubuntu/RHEL | 64 | PyTorch v1.12 + ZenDNN binary release package with python v3.10. MD5 Checksum: 4006218ed41713c5025909443a440c08 |
| PT_v1.12_ZenDNN_v4.0_C++_API.zip | 4 | 74 | 1/20/2023 | Ubuntu/RHEL | 64 | PyTorch v1.12 + ZenDNN binary release package for C++ API interface. MD5 Checksum: 3cb2aae954993f8a93aa69f36f39399a |
| PT_v1.12_ZenDNN_v4.0_source_code.tar.gz | 4 | 124 | 1/20/2023 | Ubuntu/RHEL | 64 | Integration source code of PyTorch v1.12 with ZenDNN. The instruction file inside zip code describe how to use the integration source code and how to build PyTorch v1.12 with ZenDNN. MD5 Checksum:6318232592dd0d181a29ed0f1fef09fd |
| ONNX Runtime | ||||||
| ONNXRT_v1.12.1_ZenDNN_v4.0_Python_v3.7.zip | 4 | 22 | 1/20/2023 | Ubuntu/RHEL | 64 | ONNX Runtime v1.12.1 + ZenDNN binary release package with python v3.7. MD5 Checksum: 7a809f6be3328e07e9aa45a790904825 |
| ONNXRT_v1.12.1_ZenDNN_v4.0_Python_v3.8.zip | 4 | 22 | 1/20/2023 | Ubuntu/RHEL | 64 | ONNX Runtime v1.12.1 + ZenDNN binary release package with python v3.8. MD5 Checksum: 7ef3e7dc582b05f0326a7c186d410497 |
| ONNXRT_v1.12.1_ZenDNN_v4.0_Python_v3.9.zip | 4 | 22 | 1/20/2023 | Ubuntu/RHEL | 64 | ONNX Runtime v1.12.1 + ZenDNN binary release package with python v3.9. MD5 Checksum: 79aa052d69fc36fe5f22cf6d7afbf258 |
| ONNXRT_v1.12.1_ZenDNN_v4.0_Python_v3.10.zip | 4 | 22 | 1/20/2023 | Ubuntu/RHEL | 64 | ONNX Runtime v1.12.1 + ZenDNN binary release package with python v3.10. MD5 Checksum: e167aaba7e3e8696d11a8e2f2ec2219b |
| ONNX Runtime Windows | ||||||
| ONNXRT_v1.12.1_ZenDNN_v4.0_Python_v3.7_Win.zip | 4 | 12 | 1/20/2023 | Windows | 64 | ONNX Runtime v1.12.1 + ZenDNN binary release package with python v3.7 for Windows OS. MD5 Checksum: 807113d743aa34329d5b31f5cb719881 |
| ONNXRT_v1.12.1_ZenDNN_v4.0_Python_v3.8_Win.zip | 4 | 12 | 1/20/2023 | Windows | 64 | ONNX Runtime v1.12.1 + ZenDNN binary release package with python v3.8 for Windows OS. MD5 Checksum: b2bf91c563a393451d5811ba248dfa17 |
| ONNXRT_v1.12.1_ZenDNN_v4.0_Python_v3.9_Win.zip | 4 | 12 | 1/20/2023 | Windows | 64 | ONNX Runtime v1.12.1 + ZenDNN binary release package with python v3.9 for Windows OS. MD5 Checksum: 0d7d0028d455d9fe1ffe87023dd8242e |
| ONNXRT_v1.12.1_ZenDNN_v4.0_Python_v3.10_Win.zip | 4 | 12 | 1/20/2023 | Windows | 64 | ONNX Runtime v1.12.1 + ZenDNN binary release package with python v3.10 for Windows OS. MD5 Checksum: 451cead3218693925f3a25d5021872fe |
Resources
| Version | Resources |
| 3.3 | TensorFlow-ZenDNN Plug-in User Guide |
| 3.3 | ZenDNN 3.3 – Documents |
Archives v3.3
| File Name | Version | Size | Launch Date | OS | Bitness | Description |
| TensorFlow | ||||||
| TF_v2.9_ZenDNN_v3.3_Python_v3.7.zip | 3.3 | 228 MB | 06/15/2022 | Ubuntu/RHEL | 64 | TF v2.9 + ZenDNN binary release package with python v3.7. MD5 Checksum: 9d1aa54e8a4aeb79a8688e46bc3dc235 |
| TF_v2.9_ZenDNN_v3.3_Python_v3.8.zip | 3.3 | 228 MB | 06/15/2022 | Ubuntu/RHEL | 64 | TF v2.9 + ZenDNN binary release package with python v3.8. MD5 Checksum: 75672df609140c010d4c36c7cd1ca5cf |
| TF_v2.9_ZenDNN_v3.3_Python_v3.9.zip | 3.3 | 228 MB | 06/15/2022 | Ubuntu/RHEL | 64 | TF v2.9 + ZenDNN binary release package with python v3.9. MD5 Checksum: ffbf00787378526c726fd1dbe3769c93 |
| TF_v2.9_ZenDNN_v3.3_Python_v3.10.zip | 3.3 | 228 MB | 06/15/2022 | Ubuntu/RHEL | 64 | TF v2.9 + ZenDNN binary release package with python v3.10. MD5 Checksum: af0fd75f4b76810d6ba8eb25bcb82058 |
| TF_v2.9_ZenDNN_v3.3_API.zip | 3.3 | 130 MB | 06/15/2022 | Ubuntu/RHEL | 64 | TF v2.9 + ZenDNN binary release package for C++ API interface. MD5 Checksum: 51b3b4093775ff2b67e06f18d01b41ac |
| PyTorch | ||||||
| PT_v1.11.0_ZenDNN_v3.3_Python_v3.7.zip | 3.3 | 98 MB | 06/15/2022 | Ubuntu/RHEL | 64 | PT v1.11.0 + ZenDNN binary release package with python v3.7. MD5 Checksum: 6f1da3b0420b1c1dd568fbaaae90e9ba |
| PT_v1.11.0_ZenDNN_v3.3_Python_v3.8.zip | 3.3 | 98 MB | 06/15/2022 | Ubuntu/RHEL | 64 | PT v1.11.0 + ZenDNN binary release package with python v3.8. MD5 Checksum: f76a62e9eb6e352ecc2a2ab2b68e9f38 |
| PT_v1.11.0_ZenDNN_v3.3_Python_v3.9.zip | 3.3 | 98 MB | 06/15/2022 | Ubuntu/RHEL | 64 | PT v1.11.0 + ZenDNN binary release package with python v3.9. MD5 Checksum: 318408befd3e5d099d0a1f7644c9d80d |
| PT_v1.11.0_ZenDNN_v3.3_Python_v3.10.zip | 3.3 | 98 MB | 06/15/2022 | Ubuntu/RHEL | 64 | PT v1.11.0 + ZenDNN binary release package with python v3.10. MD5 Checksum: d447a9a99d4481e92aba78a840238ed3 |
| PT_v1.11.0_ZenDNN_v3.3_API.zip | 3.3 | 56 MB | 06/15/2022 | Ubuntu/RHEL | 64 | PT v1.11.0 + ZenDNN binary release package for C++ API interface. MD5 Checksum: a191f2305f1cae6e00c82a1071df9708 |