For testing or inquiries, please visit our YouTube channel.
We provide detailed reviews and testing results for various thermal interfaces. Contact info can be found in our YouTube channel details.
Requirements
Windows 11 Intel GPU: Arc Alchemist or Battlemage dGPU. Tested on A770 16GB. ARC Driver: 32.0.101.6314 or later Miniconda (Not required but this is what I will use to setup my virtual environment)
Step 1: Create Conda Environment Named "xpu"
conda create -n xpu python=3.12 conda activate xpu
Step 2: Navigate To Data Folder
This will be where Jupyter Lab stores files In my example this is located at G:\GIT\public\XPUTEST
Step 3: Install The Following Packages
pip3 install --pre torch torchvision torchaudio torchdata safetensors matplotlib pillow numpy --index-url https://download.pytorch.org/whl/nightly/xpu pip install jupyterlab accelerate diffusers tqdm IProgress transformers jupyter lab
Step 4: Open A Test Notebook In Jupyter Lab
Execute the following code, it will take a minute to run: # This should return "True" if the ARC GPU is detected import torch torch.xpu.is_available()
Step 5: Run A Test Load Against ARC
In a new cell run the following code:
import torch
import time
matrix_size = 3000
loops = 10
torch.manual_seed(42)
torch.xpu.manual_seed(42)
device_cpu = torch.device("cpu")
A_cpu = torch.randn(matrix_size, matrix_size, device=device_cpu)
B_cpu = torch.randn(matrix_size, matrix_size, device=device_cpu)
start_time = time.time()
for _ in range(loops):
with torch.autocast(device_type="cpu", enabled=False):
mul = torch.matmul(A_cpu, B_cpu)
cpu_time = time.time() - start_time
print(f"CPU time: {cpu_time:.8f}")
torch.manual_seed(42)
torch.xpu.manual_seed(42)
device_xpu = torch.device("xpu")
A_xpu = torch.randn(matrix_size, matrix_size, device=device_xpu)
B_xpu = torch.randn(matrix_size, matrix_size, device=device_xpu)
start_time = time.time()
for _ in range(loops):
with torch.autocast(device_type="xpu", dtype=torch.float16, enabled=True):
mul = torch.matmul(A_xpu, B_xpu)
xpu_time = time.time() - start_time
print(f"XPU time: {xpu_time:.8f}")
Basic Stable Diffusion Example
torch.xpu.empty_cache()
device = torch.device("xpu" if torch.xpu.is_available() else "cpu")
from diffusers import StableDiffusionPipeline, StableDiffusionImg2ImgPipeline
model_id = "runwayml/stable-diffusion-v1-5"
stable_diffusion_txt2img = StableDiffusionPipeline.from_pretrained(model_id, use_safetensors=True,
safety_checker=None,
feature_extractor=None,
requires_safety_checker=False,)
components = stable_diffusion_txt2img.components
stable_diffusion_img2img = StableDiffusionImg2ImgPipeline(**components)
stable_diffusion_txt2img_custom = StableDiffusionPipeline(
vae=stable_diffusion_txt2img.vae,
text_encoder=stable_diffusion_txt2img.text_encoder,
tokenizer=stable_diffusion_txt2img.tokenizer,
unet=stable_diffusion_txt2img.unet,
scheduler=stable_diffusion_txt2img.scheduler,
safety_checker=None,
feature_extractor=None,
requires_safety_checker=False,
).to(device)
print(device)
import numpy as np
from IPython.display import display
import torch
prompt = "race car drifting"
negative_prompt = ""
num_inference_steps = 30
num_images_per_prompt = 1
guidance_scale = 7.5
strength = 0.7
height = 512
width = 512
# Generate an initial image
initial_image = stable_diffusion_txt2img(prompt=prompt,
strength=strength,
num_inference_steps=num_inference_steps,
guidance_scale=guidance_scale,
height=height, width=width)[0]
display(initial_image[0])
# Refine the image using the img2img pipeline
refined_image = stable_diffusion_img2img(prompt=prompt,image=initial_image,strength=strength,guidance_scale=guidance_scale,num_inference_steps=num_inference_steps,height=height,width=width)[0]
# Display the refined image
display(refined_image[0])
torch.xpu.empty_cache()
List of currently installed packages, everything works [torch torchvision/audio stable-diffusion]
(xpu) C:\Users\user>pip list Package Version ------------------------- ---------------------- accelerate 1.2.0 anyio 4.7.0 argon2-cffi 23.1.0 argon2-cffi-bindings 21.2.0 arrow 1.3.0 asttokens 3.0.0 async-lru 2.0.4 attrs 24.2.0 babel 2.16.0 beautifulsoup4 4.12.3 bleach 6.2.0 certifi 2024.8.30 cffi 1.17.1 charset-normalizer 3.3.2 colorama 0.4.6 comm 0.2.2 contourpy 1.2.1 cycler 0.12.1 debugpy 1.8.9 decorator 5.1.1 defusedxml 0.7.1 diffusers 0.31.0 executing 2.1.0 fastjsonschema 2.21.1 filelock 3.16.1 fonttools 4.53.0 fqdn 1.5.1 fsspec 2024.10.0 h11 0.14.0 httpcore 1.0.7 httpx 0.28.1 huggingface-hub 0.26.5 idna 3.10 importlib_metadata 8.5.0 intel-cmplr-lib-rt 2025.0.2 intel-cmplr-lib-ur 2025.0.2 intel-cmplr-lic-rt 2025.0.2 intel-sycl-rt 2025.0.2 IProgress 0.4 ipykernel 6.29.5 ipython 8.30.0 isoduration 20.11.0 jedi 0.19.2 Jinja2 3.1.4 json5 0.10.0 jsonpointer 3.0.0 jsonschema 4.23.0 jsonschema-specifications 2024.10.1 jupyter_client 8.6.3 jupyter_core 5.7.2 jupyter-events 0.10.0 jupyter-lsp 2.2.5 jupyter_server 2.14.2 jupyter_server_terminals 0.5.3 jupyterlab 4.3.2 jupyterlab_pygments 0.3.0 jupyterlab_server 2.27.3 kiwisolver 1.4.5 MarkupSafe 2.1.5 matplotlib 3.9.0 matplotlib-inline 0.1.7 mistune 3.0.2 mpmath 1.3.0 nbclient 0.10.1 nbconvert 7.16.4 nbformat 5.10.4 nest-asyncio 1.6.0 networkx 3.4.2 notebook_shim 0.2.4 numpy 2.1.2 overrides 7.7.0 packaging 24.1 pandocfilters 1.5.1 parso 0.8.4 pillow 11.0.0 pip 24.2 platformdirs 4.3.6 prometheus_client 0.21.1 prompt_toolkit 3.0.48 psutil 6.1.0 pure_eval 0.2.3 pycparser 2.22 Pygments 2.18.0 pyparsing 3.1.2 python-dateutil 2.9.0 python-json-logger 2.0.7 pywin32 308 pywinpty 2.0.14 PyYAML 6.0.2 pyzmq 26.2.0 referencing 0.35.1 regex 2024.11.6 requests 2.32.3 rfc3339-validator 0.1.4 rfc3986-validator 0.1.1 rpds-py 0.22.3 safetensors 0.4.5 Send2Trash 1.8.3 setuptools 75.1.0 six 1.16.0 sniffio 1.3.1 soupsieve 2.6 stack-data 0.6.3 sympy 1.13.1 tcmlib 1.2.0 terminado 0.18.1 tinycss2 1.4.0 tokenizers 0.21.0 torch 2.6.0.dev20241206+xpu torchaudio 2.5.0.dev20241206+xpu torchdata 0.10.0.dev20241206+cpu torchvision 0.20.0.dev20241206+xpu tornado 6.4.2 tqdm 4.67.1 traitlets 5.14.3 transformers 4.47.0 types-python-dateutil 2.9.0.20241206 typing_extensions 4.12.2 umf 0.9.1 uri-template 1.3.0 urllib3 2.2.3 wcwidth 0.2.13 webcolors 24.11.1 webencodings 0.5.1 websocket-client 1.8.0 wheel 0.44.0 zipp 3.21.0