Environment:Intel Ipex llm Windows Environment
| Knowledge Sources | |
|---|---|
| Domains | Infrastructure, Environment_Diagnostics |
| Last Updated | 2026-02-09 04:00 GMT |
Overview
Windows environment for IPEX-LLM with Intel GPU, supporting environment diagnostics and validation of Intel Arc GPU drivers on Windows 10/11.
Description
This environment provides a Windows-based context for running IPEX-LLM workloads on Intel GPUs. It targets Windows 10 and Windows 11 with Intel Arc GPU drivers installed. The environment supports environment diagnostics to validate that the Intel GPU is correctly detected, drivers are up to date, and Python with IPEX-LLM can access the XPU device. Python 3.9 or later is required, and the Intel GPU driver must be installed via Intel's Windows driver packages.
Usage
Use this environment for Windows Environment Diagnostics to verify Intel XPU hardware and driver readiness on Windows. It is the mandatory prerequisite for running environment validation checks and ensuring the Windows system is properly configured for IPEX-LLM inference.
System Requirements
| Category | Requirement | Notes |
|---|---|---|
| OS | Windows 10/11 (64-bit) | Latest Windows updates recommended |
| Hardware | Intel Arc GPU (A-series) | Intel Arc A770, A750, A580, or similar discrete GPU |
| GPU Driver | Intel Arc GPU drivers | Latest Intel Graphics Driver for Windows |
| Python | Python 3.9+ | Python 3.11 recommended; use Conda or official installer |
Dependencies
System Packages
- Intel Arc GPU Driver for Windows (latest version)
- Microsoft Visual C++ Redistributable (latest)
Python Packages
- `ipex-llm[xpu]` (pre-release)
- `torch` (XPU variant for Windows)
- `intel_extension_for_pytorch` (XPU variant for Windows)
Credentials
No credentials are required. The following environment setup is needed:
- Intel GPU driver must be installed via Intel Driver & Support Assistant or manual download from Intel's website.
Quick Install
# Create a Conda environment (recommended)
conda create -n ipex-llm python=3.11 -y
conda activate ipex-llm
# Install IPEX-LLM with XPU support (Windows)
pip install --pre --upgrade ipex-llm[xpu] --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/
# Verify XPU device in Python
python -c "import torch; import intel_extension_for_pytorch; print(torch.xpu.device_count(), 'XPU device(s) found')"
Common Errors
| Error Message | Cause | Solution |
|---|---|---|
| `torch.xpu.device_count() returns 0` | Intel GPU driver not installed or outdated | Install or update Intel Arc GPU driver from Intel's website |
| `No module named 'intel_extension_for_pytorch'` | IPEX not installed for Windows | Install `ipex-llm[xpu]` with the Intel PyTorch extension index |
| `Python version not supported` | Python version below 3.9 | Upgrade to Python 3.9 or later (3.11 recommended) |
| `DLL load failed` | Missing Visual C++ Redistributable | Install latest Microsoft Visual C++ Redistributable |
Compatibility Notes
- Windows Only: This environment is specific to Windows 10/11. For Linux environment diagnostics, see the Linux_XPU_Environment page.
- Conda Recommended: Using Conda for Python environment management is recommended on Windows for easier dependency resolution.
- Driver Updates: Intel GPU drivers on Windows should be kept up to date via Intel Driver & Support Assistant for best compatibility.
- No OneAPI Required: Unlike Linux, Windows environments do not require sourcing OneAPI setvars. The Intel GPU driver includes necessary runtime components.