Quickstart: Run PyTorch Inference on Intel GPU using Docker (on Linux or WSL)

[liu-shaojun]

Description

1. Why the change?

• Add a quickstart to run PyTorch Inference on Intel GPU using Docker
• Update ipex-llm-xpu image to add benchmark & examples

You may check http://10.239.44.83:8008/doc/LLM/Quickstart/docker_pytorch_inference_gpu.html to see the quickstart.

[jason-dai]

We just need one quick start for running PyTorch inference on GPU using Docker (on Linux or WSL):

1. Install docker
2. Launch docker
3. Run inference benchmark
4. Run chat.py?
5. Run PyTorch examples

[liu-shaojun]

We just need one quick start for running PyTorch inference on GPU using Docker (on Linux or WSL):

1. Install docker
2. Launch docker
3. Run inference benchmark
4. Run chat.py?
5. Run PyTorch examples

Updated