LLM benchmark for chatglm3-6b Unable to run properly

[vincent-wsz]

Run LLM benchmark for chatglm3-6b , prompt“OSError: [WinError 126] 找不到指定的模块。 Error loading "D:\miniforge3\envs\llm\Lib\site-packages\intel_extension_for_pytorch\bin\intel-ext-pt-gpu.dll" or one of its dependencies.” Test environment：Ultra 5 125H CPU，Win11 23H2 Pro，gfx driver-32.0.101.5972 These are the pip list in my container: Installation Steps：

1. install Miniforge3-Windows-x86_64.exe，VSCodeUserSetup-x64-1.89.1.exe，w_BaseKit_p_2024.1.0.595.exe
2. conda create -n llm python=3.10 libuv
3. conda activate llm
4. pip install --pre --upgrade ipex-llm[xpu] --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/cn/ pip install torch==2.1.0.post3 torchvision==0.16.0.post3 torchaudio==2.1.0.post3 intel-extension-for-pytorch==2.1.40+xpu --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/mtl/cn/ pip install transformers==4.38.2 pip install omegaconf pandas
5. cd D:\ipex-handson\script\all-in-one
6. set SYCL_CACHE_PERSISTENT=1 set BIGDL_LLM_XMX_DISABLED=1
7. 修改config.yaml
8. python run.py

[Oscilloscope98]

Hi @vincent-wsz,

Your ipex and oneAPI version seems to be wrong for ipex-llm GPU support on PyTorch 2.1.

Please refer to our GPU installation guide on Windows for more information :)