# Can’t switch to GPU Hardware Upgrade

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcan-t-switch-to-gpu-hardware-upgrade%2F28759%3E%5C
**Category:** Spaces\
**Created:** [January 1, 2023, 11:30pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcan-t-switch-to-gpu-hardware-upgrade%2F28759 "2023-01-01T23:30:44Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![Testeicu](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/testeicu/32/13304_2.png) [@Testeicu](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FTesteicu)%5C
**Post date:** [January 1, 2023, 11:30pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcan-t-switch-to-gpu-hardware-upgrade%2F28759%2F1 "2023-01-01T23:30:44Z")

</div>

Hey,

I want to upgrade one of my space to upgrade with a GPU. I switch to the „Settings tab“ and select a T4 small, it triggers a new building process but it remains in this state. I have try it with a few duplicated Space however all hab the same problem.

Any Ideas?

 ![image](https://us1.discourse-cdn.com/hellohellohello/original/2X/e/ebfae11592398a4510ffc98c69d1dd5c9e656dd5.jpeg)

---

<div class="post-metadata">

**Author:** ![fjenett](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/fjenett/32/59016_2.png) [@fjenett](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Ffjenett)%5C
**Post date:** [January 2, 2023, 12:47pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcan-t-switch-to-gpu-hardware-upgrade%2F28759%2F2 "2023-01-02T12:47:31Z")

</div>

Same for me trying to launch a Docker based space. GPU-backed spaces seem to hang at application startup (“Scheduling Space”). Maybe there is not enough GPU at the moment?

---

<div class="post-metadata">

**Author:** ![chris-rannou](https://avatars.discourse-cdn.com/v4/letter/c/7feea3/32.png) [@chris-rannou](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fchris-rannou)%5C
**Post date:** [January 5, 2023, 3:54pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcan-t-switch-to-gpu-hardware-upgrade%2F28759%2F3 "2023-01-05T15:54:39Z")

</div>

GPUs infra was indeed a bit strained at the time. GPUs should be available now

---

<div class="post-metadata">

**Author:** ![leyne](https://avatars.discourse-cdn.com/v4/letter/l/858c86/32.png) [@leyne](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fleyne)%5C
**Post date:** [January 21, 2023, 11:04am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcan-t-switch-to-gpu-hardware-upgrade%2F28759%2F4 "2023-01-21T11:04:51Z")

</div>

Hey there, I have been suffering from the same “Scheduling Space” hanging problem for the entire day. Is this an issue with HuggingFace service, or is there something I should be doing to solve this problem? It eventually times out and restarts the build.

---

<div class="post-metadata">

**Author:** ![leyne](https://avatars.discourse-cdn.com/v4/letter/l/858c86/32.png) [@leyne](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fleyne)%5C
**Post date:** [January 23, 2023, 8:36pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcan-t-switch-to-gpu-hardware-upgrade%2F28759%2F5 "2023-01-23T20:36:00Z")

</div>

For anyone else out there that is encountering a similar issue, someone from Hugging Face reached out to me to let me know that I can try buffing up the compute I was using to solve the issue. This was resolved when I went from A10G small to T4 medium, which increased my RAM from 15GB to 30GB.
