Skip to content
This repository was archived by the owner on Jun 24, 2024. It is now read-only.
This repository was archived by the owner on Jun 24, 2024. It is now read-only.

Free tensors from RAM if they are offloaded to an Accelerator #390

Open
Open
Free tensors from RAM if they are offloaded to an Accelerator#390
Labels
issue:enhancementNew feature or requesttopic:clblasthttps://github.com/CNugteren/CLBlast supporttopic:cublashttps://developer.nvidia.com/cublas supporttopic:metalhttps://developer.apple.com/metal/ support
@LLukas22

Description

@LLukas22
LLukas22
opened on Jul 27, 2023

Right now the data of a tensor isn't freed if it is offloaded to a GPU. We should fix that to enable users to run bigger models which are split between CPU and GPU.

Metadata

Metadata

Assignees

No one assigned

    Labels

    issue:enhancementNew feature or requesttopic:clblasthttps://github.com/CNugteren/CLBlast supporttopic:cublashttps://developer.nvidia.com/cublas supporttopic:metalhttps://developer.apple.com/metal/ support

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions