intel/intel-extension-for-pytorch issues and pull requests

#713 - When using xpu backend, stable diffusion give bad result (comparing with 'cpu' backend and also opevino)

Issue - State: open - Opened by LeptonWu 7 days ago - 1 comment

#712 - No XPU devices found

Issue - State: open - Opened by Bhargav230m 8 days ago - 5 comments

#711 - iGPU in pentium G4620 generates bad image by diffusers when setting data type to bfloat16

Issue - State: closed - Opened by LeptonWu 8 days ago - 5 comments

#710 - OSError: [WinError 193] %1 is not a valid win32 application. Error loading "...\intel-ext-pt-gpu.dll" or one of its dependencies.

Issue - State: open - Opened by 0Pinky0 8 days ago - 2 comments

#709 - Performance Regression?

Issue - State: open - Opened by chsasank 9 days ago - 1 comment

#708 - Sanity Check Failed on A770

Issue - State: open - Opened by chunsheong 9 days ago - 4 comments

#707 - DataParallel is Supported for XPU?

Issue - State: open - Opened by yash3056 11 days ago - 4 comments

#706 - `intel-extension-for-pytorch=2.3.110` package not available

Issue - State: open - Opened by idkSeth 12 days ago - 1 comment

#705 - NotImplementedError: Could not run 'aten::_upsample_bicubic2d_aa.out' with arguments from the 'XPU' backend.

Issue - State: open - Opened by Nuullll 15 days ago
Labels: ARC, Escalate

#704 - unable to install pytorch and ipex

Issue - State: open - Opened by nishtsha 16 days ago - 1 comment
Labels: ARC, Installation

#703 - Facing Segmentation Fault on MaskedMHA Test

Issue - State: open - Opened by Alok-Ranjan23 16 days ago
Labels: LLM

#702 - Can't find device after installing intel pytorch 2.3.1

Issue - State: closed - Opened by bisjack 18 days ago - 16 comments
Labels: ARC, Installation

#701 - Run finetune.py on Xeon but failed - no attribute "weight"

Issue - State: open - Opened by JamieVC 18 days ago - 7 comments
Labels: CPU, LLM, Escalate

#700 - Can't import torch

Issue - State: open - Opened by AlphastT101 19 days ago - 5 comments
Labels: ARC, Windows, Installation

#699 - Required aspect fp64 is not supported on the device when running cross_entropy

Issue - State: open - Opened by YumingChang02 21 days ago - 3 comments
Labels: XPU/GPU, ARC

#698 - loss.backward() on an LSTM causes SegFault on Arc GPU

Issue - State: open - Opened by LivesayMe 24 days ago - 5 comments
Labels: XPU/GPU, ARC, Crash

#697 - conda repo has outdated repodata.bz2, there is no `intel-extension-for-pytorch`

Issue - State: open - Opened by chisaato 26 days ago - 3 comments
Labels: XPU/GPU, Functionality, Installation

#696 - torch.xpu for GQA

Issue - State: open - Opened by liuxingbin about 1 month ago - 6 comments
Labels: XPU/GPU, Query

#695 - Can you provide me with a prompt.json file template

Issue - State: open - Opened by zwx109473 about 1 month ago - 1 comment
Labels: XPU/GPU, Query

#694 - Update bfloat16 conversion property

Pull Request - State: closed - Opened by anmyachev about 1 month ago - 1 comment

#693 - Building with torch nightly (torch 2.5) for XPU

Issue - State: open - Opened by narendrachaudhary51 about 1 month ago - 8 comments
Labels: XPU/GPU, Query

#692 - The training code provided in intel-extension-for-pytorch example still not working

Issue - State: closed - Opened by Jaja8877 about 1 month ago

#691 - can't use ComfyUI_examples's FLUX fp8 workflow in ipex=2.1.30 and 2.1.40

Issue - State: closed - Opened by kingsdom-sk about 1 month ago - 8 comments
Labels: XPU/GPU, Windows

#690 - docker: Update prebuilt Dockerfile

Pull Request - State: open - Opened by ChengyuZhu6 about 1 month ago

#689 - Powershell environment scripts to fix "[WinError 126] The specified module could not be found. Error loading "C:\path\to\venv\lib\site-packages\torch\lib\backend_with_compiler.dll" or one of its dependencies"

Issue - State: closed - Opened by han-minhee about 1 month ago - 5 comments
Labels: XPU/GPU, Windows

#688 - 0 <= device.index() && device.index() < static_cast<c10::DeviceIndex>(device_ready_queues_.size()) INTERNAL ASSERT FAILED at "C:\\Jenkins\\workspace\\IPEX-GPU-ARC770-Windows-Build\\frameworks.ai.pytorch.private-gpu\\torch\\csrc\\autograd\\engine.cpp":1421

Issue - State: closed - Opened by niuhongbin1 about 2 months ago - 4 comments
Labels: NotAnIssue

#687 - When I try to run the fp8 quantized flux model with ComfyUI, the following problem occurs:

Issue - State: closed - Opened by 1ngram433 about 2 months ago - 8 comments
Labels: Bug, XPU/GPU, ARC

#686 - How to save and load ipex optimized model?

Issue - State: open - Opened by benja-matic about 2 months ago - 12 comments
Labels: Documentation, CPU, NotAnIssue, LLM

#685 - Update device property names

Pull Request - State: closed - Opened by ZzEeKkAa about 2 months ago - 1 comment

#684 - ipex.optimize removes attached hooks to the model

Issue - State: closed - Opened by karnikkanojia about 2 months ago - 5 comments
Labels: CPU, Crash

#683 - can not run pyannote audio on intel a770

Issue - State: closed - Opened by SlyRebula about 2 months ago - 3 comments
Labels: ARC

#682 - torch scatter operation return wrong result on xpu.

Issue - State: closed - Opened by qiuxin2012 about 2 months ago - 2 comments
Labels: ARC, Correctness

#681 - Trying llama3.1 on Xeon and getting rope scaling error

Issue - State: closed - Opened by js333031 2 months ago - 10 comments
Labels: CPU, Query

#680 - tensor parallel support for multi intel GPU

Issue - State: closed - Opened by bopeng1234 2 months ago - 7 comments
Labels: XPU/GPU, Query

#679 - How to use Arc GPU in Pytorch Lightning ?

Issue - State: open - Opened by 771991673 2 months ago - 1 comment
Labels: XPU/GPU, Query

#678 - Intel PyTorch extension doesn't detect Data Center Max GPU

Issue - State: closed - Opened by saiprasadb01 2 months ago - 5 comments
Labels: XPU/GPU, NotAnIssue

#677 - Irregular numbers on IPEX 2.3.0 and 2.3.100 for Bert-Large and Distil-bert

Issue - State: open - Opened by ramaseshan1 2 months ago - 7 comments
Labels: CPU, Performance

#676 - Questions about the eager mode

Issue - State: closed - Opened by zhangwm-pt 3 months ago - 8 comments
Labels: Query

#675 - Failed to generate LLM docker container

Issue - State: closed - Opened by jingzhaoou 3 months ago - 3 comments
Labels: Bug, CPU, dependencies

#674 - [Windows] [Pyinstaller] Process exit when I use "Pyinstaller" to package, the executable cannot detect XPU.

Issue - State: closed - Opened by tjtanaa 3 months ago - 14 comments
Labels: XPU/GPU, Windows

#673 - RuntimeError: "dot" not implemented for 'Half'

Issue - State: closed - Opened by NineMeowICT 3 months ago - 5 comments
Labels: XPU/GPU, ARC, Crash, Functionality

#672 - RuntimeError: XPU out of memory.

Issue - State: closed - Opened by rohitpreddy07 3 months ago - 9 comments
Labels: XPU/GPU, Crash

#671 - DPC++ Extension is not enabled on Windows OS

Issue - State: closed - Opened by YixinSong-e 3 months ago - 6 comments

#670 - Unable to run LanguageBind/Video-LLaVA-7B-hf using ipex

Issue - State: closed - Opened by shailesh837 3 months ago - 3 comments
Labels: ARC, Crash

#669 - Running IPEX samples - environment setup clarification

Issue - State: closed - Opened by js333031 3 months ago - 3 comments
Labels: CPU, LLM

#668 - `get_compiler_abi_compatibility_and_version` cannot parse version string with "git" suffix

Issue - State: closed - Opened by benjaminum 3 months ago - 6 comments
Labels: Bug, XPU/GPU, Escalate

#667 - Check extension supportance via `ext_oneapi_supports_cl_extension`

Pull Request - State: closed - Opened by LiyangLingIntel 3 months ago - 1 comment

#666 - ImportError: libze_loader.so.1: cannot open shared object file: No such file or directory

Issue - State: closed - Opened by bconsolvo 3 months ago - 3 comments
Labels: XPU/GPU, Crash

#665 - OSError: [WinError 126] The specified module could not be found. Error loading "\.venv\Lib\site-packages\torch\lib\fbgemm.dll" or one of its dependencies

Issue - State: closed - Opened by Tisha-linkenite 3 months ago - 5 comments
Labels: XPU/GPU, Windows

#664 - does ipex.llm support speculative sampling?

Issue - State: closed - Opened by zhangnju 3 months ago - 1 comment
Labels: Query, LLM

#663 - ipex.optimize without setting replace_dropout_with_identity to False errors out when trying to run Pixart Sigma ComfyUI on IPEX 2.1.30+xpu

Issue - State: closed - Opened by simonlui 3 months ago - 15 comments
Labels: XPU/GPU, Crash, NotAnIssue

#662 - [torch]: ImportError: cannot import name 'packaging' from 'pkg_resources'

Issue - State: closed - Opened by raj-ritu17 3 months ago - 4 comments
Labels: NotAnIssue

#661 - how to free gpu without restarting python

Issue - State: closed - Opened by 13917904019 3 months ago - 2 comments
Labels: NotAnIssue, dependencies

#660 - "cannot import name 'BUFSIZE' from 'numpy'" - traceback from "Sanity Check"

Issue - State: closed - Opened by BrianSwift-Intel 3 months ago - 5 comments
Labels: XPU/GPU, NotAnIssue

#659 - Compilation Intel Extension For PyTorch with specific AOT device enabled didn't reduce AOT for first pass tensor to model.

Issue - State: closed - Opened by MaksimDanilov 3 months ago - 8 comments
Labels: iGPU, Performance, Windows

#658 - Request support for pytorch lighting and provide installation packages for deepspeed

Issue - State: closed - Opened by uniartisan 4 months ago - 12 comments
Labels: XPU/GPU, Query

#657 - Doubt: Does ipex.llm.optimize() perform any optimization in run_quantization.py when "qconfig_summary_file" is given

Issue - State: closed - Opened by MadhumithaSrini 4 months ago - 2 comments
Labels: Query

#656 - Torchvision: reading an image does not work.

Issue - State: closed - Opened by RafinRono 4 months ago - 3 comments
Labels: ARC, iGPU, Windows

#655 - OSerror during run : how do I config with huggingface before run

Issue - State: closed - Opened by pauldintel 4 months ago - 2 comments
Labels: UserExperience, NotAnIssue

#654 - [CI] enable code format check

Pull Request - State: closed - Opened by DiweiSun 4 months ago

#653 - OSError When Importing Pytorch

Issue - State: closed - Opened by MikyLanfra 4 months ago - 9 comments
Labels: XPU/GPU, UserExperience, iGPU, Windows

#652 - bad spirv bitcode: spirv_read_verify: verify Module failed Build failed with error code: -11

Issue - State: closed - Opened by maxwell-zhengxu 4 months ago - 3 comments
Labels: ARC, Compilation

#651 - Import Error: "libmkl_sycl_blas.so.4" and "libze_loader.so.1" cannot open shared object file: No such file or directory.

Issue - State: closed - Opened by brendondgr 4 months ago - 18 comments

#650 - IPEX Fusions for BF16

Issue - State: closed - Opened by sreelekshmyanil 4 months ago - 5 comments
Labels: Query

#649 - Load model trained in XPU and fails to continue training

Issue - State: closed - Opened by hermanhmchan 4 months ago - 6 comments
Labels: Documentation, ARC, NotAnIssue

#648 - Qwen-7b int8 inference fails when using run_quantization.py script

Issue - State: closed - Opened by MadhumithaSrini 4 months ago - 6 comments

#647 - reduce_scatter raises a RuntimeError

Issue - State: closed - Opened by garrett361 4 months ago - 9 comments

#646 - FSDP raises ZE_RESULT_ERROR_OUT_OF_DEVICE_MEMORY in multi-node usage

Issue - State: closed - Opened by garrett361 4 months ago - 13 comments
Labels: Bug, dGPU-Max

#645 - Fail to run LLM inference with ipex 2.1.30 and oneapi 2024.1 on Windows iGPU

Issue - State: closed - Opened by plusbang 4 months ago - 4 comments

#644 - RuntimeError: FFT_INVALID_DESCRIPTOR while using ChatTTS

Issue - State: closed - Opened by 771991673 4 months ago - 13 comments
Labels: XPU/GPU, Crash

#643 - Native API failed. Native API returns: -2 (PI_ERROR_DEVICE_NOT_AVAILABLE) -2 (PI_ERROR_DEVICE_NOT_AVAILABLE)

Issue - State: open - Opened by MaksimDanilov 4 months ago - 21 comments
Labels: XPU/GPU, Windows

#642 - Do I need to install Intel oneAPI base toolkit separately for IPEX?

Issue - State: closed - Opened by RafinRono 4 months ago

#641 - Will intel extension for pytorch support dual a770 in one linux PC?

Issue - State: closed - Opened by david-strejc 4 months ago - 2 comments
Labels: ARC, Query

#640 - reduce_scatter_tensor raises ZE_RESULT_ERROR_OUT_OF_DEVICE_MEMORY in multi-node usage

Issue - State: closed - Opened by garrett361 4 months ago - 34 comments
Labels: Bug, XPU/GPU

#639 - Torchaudio

Issue - State: closed - Opened by ElliottDyson 4 months ago - 1 comment
Labels: ARC, Query

#638 - nvidia/amd gpu support for ipex

Issue - State: closed - Opened by bopeng1234 4 months ago - 4 comments
Labels: CPU, Query

#637 - bfloat16 is slower than float32 on Intel Xeon Platinum 8481C CPU

Issue - State: open - Opened by netw0rkf10w 4 months ago - 19 comments
Labels: CPU, Performance

#636 - Preview changes in ComfyUI causes segmentation fault and etc. with IPEX v2.1.30+xpu

Issue - State: closed - Opened by simonlui 4 months ago - 16 comments

#635 - Can a model quantized using IPEX be run with torch.compile enabled?

Issue - State: closed - Opened by cfRod 4 months ago - 12 comments
Labels: CPU, Query, Ecosystem

#634 - `ERRORLEVEL` not equal to 0 when moving an tensor to XPU on Windows

Issue - State: closed - Opened by Oscilloscope98 4 months ago - 3 comments
Labels: Bug, ARC, iGPU, Windows

#633 - crash when out token length is > 64

Issue - State: open - Opened by Edward-Lin 4 months ago - 4 comments
Labels: ARC, Crash, Windows, LLM

#632 - Error loading "{dll}" or one of its dependencies after following windows installation instructions

Issue - State: closed - Opened by zippererer 4 months ago - 3 comments
Labels: Windows

#631 - [PyTorch-XPU] NotImplementedError: No registered fallback function for aten::view

Issue - State: closed - Opened by uniartisan 4 months ago - 7 comments
Labels: XPU/GPU, ARC, Crash

#630 - failed to apply concat_linear on unet

Issue - State: closed - Opened by nihilityer 4 months ago - 3 comments
Labels: ARC, Windows

#629 - [WSL]"XPU out of memory" error when using to("xpu") method in Intel PyTorch Extension (IPEX)

Issue - State: closed - Opened by uniartisan 4 months ago - 6 comments

#628 - GPT-J benchmarking fails with IPEX 2.2.0/2.3.0

Issue - State: closed - Opened by MadhumithaSrini 5 months ago - 9 comments
Labels: CPU, Crash, LLM

#627 - [CI] Enable code format check in CI based on Github action

Pull Request - State: closed - Opened by DiweiSun 5 months ago

#626 - Memory usage is low

Issue - State: closed - Opened by QiXuanWang 5 months ago - 10 comments
Labels: ARC, Performance

#625 - no XPU device found with IPEX-v2.1.30-xpu and oneAPI2024.1 releases running with MAX-1550

Issue - State: closed - Opened by zhenhanintel 5 months ago - 2 comments
Labels: Bug, dGPU-Max

#624 - When is ipex available for pytorch 2.3.0 ?

Issue - State: closed - Opened by sriharshaj70 5 months ago - 5 comments
Labels: Query

#623 - NotImplementedError: The following operation failed in the TorchScript interpreter. Traceback of TorchScript (most recent call last): RuntimeError: Could not run 'torch_ipex::mul_add' with arguments from the 'CPU' backend.

Issue - State: closed - Opened by wangdq1989 5 months ago - 3 comments
Labels: ARC, iGPU, Crash, Windows

GitHub / intel/intel-extension-for-pytorch issues and pull requests