sanchitintel

Hi @tedliosu, thanks for checking it out! While the link you provided does not have the latest info (it doesn't even list discrete GPUs), you're right that you might have...

Awesome! Thanks, @tedliosu! Looks like your iGPU is indeed `tgllp`! :)

Hi @tedliosu! Thanks again for the info! We'll investigate this issue while enabling `Intel Extension for PyTorch` for iGPUs. iGPUs are currently unsupported.

Hi @jingxu10, please advise if we should leave this issue open so that it'd be a known issue when/if we'd start working on iGPU support. Thanks

Hi @sterrettm2, the CI check that failed is not usually run before a PR is merged unless `ciflow/slow` label is added. I mean the failure is legit.

@pytorchbot rebase -b main

Hi @sterrettm2, you're right! That failure is unrelated & [also occurred in some other PRs](https://hud.pytorch.org/failure?name=slow%20%2F%20linux-focal-py3.9-clang10%20%2F%20test%20(slow%2C%201%2C%202%2C%20lf.linux.2xlarge)&jobName=undefined&failureCaptures=%5B%22inductor%2Ftest_cpu_repro.py%3A%3ACPUReproTests%3A%3Atest_lstm_packed%22%5D). We still need approval of core reviewer(s) since my approval won't suffice for landing.

@pytorchbot merge

@pytorchbot merge

@yanbing-j, with these changes, I observed different behavior from [earlier](https://github.com/pytorch/pytorch/issues/134449#issuecomment-2318887128) while running `generate.py`. Not sure if it's because of this PR, or because of other changes introduced in `torchchat`. With...