[WinError 126] Error loading "backend_with_compiler.dll" Intel ARC

[Magentatl]

Describe the bug

Hi all, I still meeting with problem like in thread https://github.com/oobabooga/text-generation-webui/issues/5123 - OSError: [WinError 126] The specified module could not be found. Error loading "backend_with_compiler.dll" I thought that this issue was fixed about half year ago. I tried to fix it with my no skilled hands with no success. May I ask someone to explain how to solve it?

I know that my problem may actually be minor or solved quite simply, but I've lost all hope of fixing it without anyone's help.

PS: I succeed to run it on transformers loader with a combination of guides "Install IPEX-LLM" and "Run Text Generation WebUI" with a 3.1 t/s speed (LLama 3). But I see that it is not the final solution of a problem.

Is there an existing issue for this?

• [X] I have searched the existing issues

Reproduction

A. Install latest drivers for ARC 770. Do a clean install of OobaBooga using the latest snapshot. Select "D" for the Arc install. Run start_windows.bat

Screenshot

No response

Logs

┌──────────────────────────────────────────────────────────────────────────────────── Traceback (most recent call last) ─────────────────────────────────────────────────────────────────────────────────────┐
│ C:\Users\asdas\Desktop\text-generation-webui-main\server.py:6 in <module>                                                                                                                                  │
│                                                                                                                                                                                                            │
│     5                                                                                                                                                                                                      │
│ >   6 import accelerate  # This early import makes Intel GPUs happy                                                                                                                                        │
│     7                                                                                                                                                                                                      │
│                                                                                                                                                                                                            │
│ C:\Users\asdas\Desktop\text-generation-webui-main\installer_files\env\Lib\site-packages\accelerate\__init__.py:16 in <module>                                                                              │
│                                                                                                                                                                                                            │
│   15                                                                                                                                                                                                       │
│ > 16 from .accelerator import Accelerator                                                                                                                                                                  │
│   17 from .big_modeling import (                                                                                                                                                                           │
│                                                                                                                                                                                                            │
│ C:\Users\asdas\Desktop\text-generation-webui-main\installer_files\env\Lib\site-packages\accelerate\accelerator.py:32 in <module>                                                                           │
│                                                                                                                                                                                                            │
│     31                                                                                                                                                                                                     │
│ >   32 import torch                                                                                                                                                                                        │
│     33 import torch.utils.hooks as hooks                                                                                                                                                                   │
│                                                                                                                                                                                                            │
│ C:\Users\asdas\Desktop\text-generation-webui-main\installer_files\env\Lib\site-packages\torch\__init__.py:139 in <module>                                                                                  │
│                                                                                                                                                                                                            │
│    138                 err.strerror += f' Error loading "{dll}" or one of its dependencies.'                                                                                                               │
│ >  139                 raise err                                                                                                                                                                           │
│    140                                                                                                                                                                                                     │
└────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────┘
OSError: [WinError 126] Error loading "C:\Users\asdas\Desktop\text-generation-webui-main\installer_files\env\Lib\site-packages\torch\lib\backend_with_compiler.dll" or one of its
dependencies.
dependencies.
Press any key to continue . . .

System Info

ARC A770, win 10, latest drivers 32.0.101.5762

[m0r6aN]

I am having the same problem.

[Geiokami]

same issue here! after version 1.9 i think ive been getting this error and havent found any solution.

[TinJon06]

I think this due to HTTP 403 response thus not installing the required oneAPI runtime libraries.

start_windows.bat:

*******************************************************************
* Installing Intel oneAPI runtime libraries.
*******************************************************************

Collecting package metadata (current_repodata.json): failed

UnavailableInvalidChannel: HTTP 403 FORBIDDEN for channel intel <https://conda.anaconda.org/intel>

The channel is not accessible or is invalid.

You will need to adjust your conda configuration to proceed.
Use `conda config --show channels` to view your configuration's current state,
and use `conda config --show-sources` to view config file locations.

No oneAPI, no IPEX. The correct link or channel must be updated in the installer.

[schorschie]

I changed the conda channel and added a pull request: #6433

[boif]

i have the same problem