Using IPEX community pipeline in stable-diffusion-inference-intel.md

[yingjie-han]

Currently, we add a new community pipeline to enable IPEX acceleration for stable diffusion in latest Diffusers [v0.17.0.] [Community Pipelines]Accelerate inference of stable diffusion by IPEX on CPU (https://github.com/huggingface/diffusers/pull/3105)

Different from the IPEX optimized methods already posted in this blog, the new community pipeline can benefit more from TorchScript mode optimization which enables MHA optimization with Flash Attention in IPEX.

To get the best performance, Diffusers v0.17.0, Pytorch/IPEX v2.0 are needed.

So we would like to ask for an update to this blog to show the latest IPEX optimization performance to community.

@juliensimon ,@echarlaix As we don't have the exactly same hardware as blog's, may I ask you for help to test the IPEX community pipeline and update the performance data?

Thanks a lot!