-
Notifications
You must be signed in to change notification settings - Fork 25.5k
Open
Labels
module: xpuIntel XPU related issuesIntel XPU related issuestriagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate module
Description
🚀 The feature, motivation and pitch
It would be nice if PyTorch added support for torch.xpu.memory.mem_get_info()
! When calling this in PyTorch 2.8.0+xpu, the following RuntimeError
shows up:
RuntimeError: The device (Intel(R) Arc(TM) Graphics) doesn't support querying the available free memory. You can file an issue at https://github.com/pytorch/pytorch/issues to help us prioritize its implementation.
This way, code (and other libraries/frameworks) can get memory information for Intel GPUs, which can be useful.
Alternatives
No response
Additional context
I was trying to use bitsandbytes
to quantize LLM for HuggingFace Transformers inference, but it required accessing memory information from the xpu
.
nicolasramy and Kevin-Delnoije
Metadata
Metadata
Assignees
Labels
module: xpuIntel XPU related issuesIntel XPU related issuestriagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate module