Skip to content

Commit 28ca4dd

Browse files
update get start xpu document for v2.7 (pytorch#150633)
update get start xpu document for v2.7 (pytorch#150397) update get start xpu document for v2.7 Pull Request resolved: pytorch#150397 Approved by: https://github.com/guangyey, https://github.com/EikanWang, https://github.com/atalman Co-authored-by: Nikita Shulga <[email protected]> (cherry picked from commit 96f35f5) Co-authored-by: ZhaoqiongZ <[email protected]>
1 parent 06c6a81 commit 28ca4dd

File tree

1 file changed

+35
-16
lines changed

1 file changed

+35
-16
lines changed

docs/source/notes/get_start_xpu.rst

Lines changed: 35 additions & 16 deletions
Original file line numberDiff line numberDiff line change
@@ -4,27 +4,46 @@ Getting Started on Intel GPU
44
Hardware Prerequisite
55
---------------------
66

7+
For Intel Data Center GPU
8+
79
.. list-table::
8-
:widths: 50 50
10+
:widths: 50 50 50 50
911
:header-rows: 1
1012

11-
* - Supported OS
12-
- Validated Hardware
13-
* - Linux
14-
- Intel® Client GPUs / Intel® Data Center GPU Max Series
15-
* - Windows
16-
- Intel® Client GPUs
17-
* - WSL2 (experimental feature)
18-
- Intel® Client GPUs
19-
20-
Intel GPUs support (Prototype) is ready in PyTorch* 2.6 for Intel® Client GPUs and Intel® Data Center GPU Max Series on both Linux and Windows, which brings Intel GPUs and the SYCL* software stack into the official PyTorch stack with consistent user experience to embrace more AI application scenarios.
13+
* - Device
14+
- Red Hat* Enterprise Linux* 9.2
15+
- SUSE Linux Enterprise Server* 15 SP5
16+
- Ubuntu* Server 22.04 (>= 5.15 LTS kernel)
17+
* - Intel® Data Center GPU Max Series (CodeName: Ponte Vecchio)
18+
- yes
19+
- yes
20+
- yes
21+
22+
For Intel Client GPU
23+
24+
+-------------------------------------+----------------------------------------------------------------------------------------------+
25+
| Supported OS | Validated Hardware |
26+
+=====================================+==============================================================================================+
27+
|| Windows 10/11 & Ubuntu 24.10 || Intel® Arc A-Series Graphics (CodeName: Alchemist) |
28+
|| || Intel® Arc B-Series Graphics (CodeName: Battlemage) |
29+
|| || Intel® Core™ Ultra Processors with Intel® Arc™ Graphics (CodeName: Meteor Lake) |
30+
|| || Intel® Core™ Ultra 200V Series with Intel® Arc™ Graphics (CodeName: Lunar Lake) |
31+
|| || Intel® Core™ Ultra Series 2 Processors with Intel® Arc™ Graphics (CodeName: Arrow Lake) |
32+
+-------------------------------------+----------------------------------------------------------------------------------------------+
33+
|| Ubuntu 24.04 & WSL2 (Ubuntu 24.04) || Intel® Arc A-Series Graphics (CodeName: Alchemist) |
34+
|| || Intel® Core™ Ultra Processors with Intel® Arc™ Graphics (CodeName: Meteor Lake) |
35+
|| || Intel® Core™ Ultra 200V Series with Intel® Arc™ Graphics (CodeName: Lunar Lake) |
36+
|| || Intel® Core™ Ultra Series 2 Processors with Intel® Arc™ Graphics (CodeName: Arrow Lake) |
37+
+-------------------------------------+----------------------------------------------------------------------------------------------+
38+
39+
Intel GPUs support (Prototype) is ready from PyTorch* 2.5 for Intel® Client GPUs and Intel® Data Center GPU Max Series on both Linux and Windows, which brings Intel GPUs and the SYCL* software stack into the official PyTorch stack with consistent user experience to embrace more AI application scenarios.
2140

2241
Software Prerequisite
2342
---------------------
2443

25-
To use PyTorch on Intel GPUs, you need to install the Intel GPUs driver first. For installation guide, visit `Intel GPUs Driver Installation <https://www.intel.com/content/www/us/en/developer/articles/tool/pytorch-prerequisites-for-intel-gpu/2-6.html#driver-installation>`_.
44+
To use PyTorch on Intel GPUs, you need to install the Intel GPUs driver first. For installation guide, visit `Intel GPUs Driver Installation <https://www.intel.com/content/www/us/en/developer/articles/tool/pytorch-prerequisites-for-intel-gpu.html#driver-installation>`_.
2645

27-
Please skip the Intel® Deep Learning Essentials installation section if you install from binaries. For building from source, please refer to `PyTorch Installation Prerequisites for Intel GPUs <https://www.intel.com/content/www/us/en/developer/articles/tool/pytorch-prerequisites-for-intel-gpu/2-6.html>`_ for both Intel GPU Driver and Intel® Deep Learning Essentials Installation.
46+
Please skip the Intel® Deep Learning Essentials installation section if you install from binaries. For building from source, please refer to `PyTorch Installation Prerequisites for Intel GPUs <https://www.intel.com/content/www/us/en/developer/articles/tool/pytorch-prerequisites-for-intel-gpu.html>`_ for both Intel GPU Driver and Intel® Deep Learning Essentials Installation.
2847

2948

3049
Installation
@@ -33,7 +52,7 @@ Installation
3352
Binaries
3453
^^^^^^^^
3554

36-
Now that we have `Intel GPU Driver <https://www.intel.com/content/www/us/en/developer/articles/tool/pytorch-prerequisites-for-intel-gpu/2-6.html#driver-installation>`_ installed, use the following commands to install ``pytorch``, ``torchvision``, ``torchaudio`` on Linux.
55+
Now that we have `Intel GPU Driver <https://www.intel.com/content/www/us/en/developer/articles/tool/pytorch-prerequisites-for-intel-gpu.html#driver-installation>`_ installed, use the following commands to install ``pytorch``, ``torchvision``, ``torchaudio`` on Linux.
3756

3857
For release wheels
3958

@@ -52,7 +71,7 @@ For nightly wheels
5271
From Source
5372
^^^^^^^^^^^
5473

55-
Now that we have `Intel GPU Driver and Intel® Deep Learning Essentials <https://www.intel.com/content/www/us/en/developer/articles/tool/pytorch-prerequisites-for-intel-gpu/2-6.html>`_ installed. Follow guides to build ``pytorch``, ``torchvision``, ``torchaudio`` from source.
74+
Now that we have `Intel GPU Driver and Intel® Deep Learning Essentials <https://www.intel.com/content/www/us/en/developer/articles/tool/pytorch-prerequisites-for-intel-gpu.html>`_ installed. Follow guides to build ``pytorch``, ``torchvision``, ``torchaudio`` from source.
5675

5776
Build from source for ``torch`` refer to `PyTorch Installation Build from source <https://github.com/pytorch/pytorch?tab=readme-ov-file#from-source>`_.
5877

@@ -88,7 +107,7 @@ If you are migrating code from ``cuda``, you would change references from ``cuda
88107
The following points outline the support and limitations for PyTorch with Intel GPU:
89108

90109
#. Both training and inference workflows are supported.
91-
#. Both eager mode and ``torch.compile`` is supported.
110+
#. Both eager mode and ``torch.compile`` is supported. The feature ``torch.compile`` is also supported on Windows from PyTorch* 2.7 with Intel GPU, refer to `How to Use Inductor on Windows with CPU/XPU <https://pytorch.org/tutorials/prototype/inductor_windows_cpu.html>`_.
92111
#. Data types such as FP32, BF16, FP16, and Automatic Mixed Precision (AMP) are all supported.
93112

94113
Examples

0 commit comments

Comments
 (0)