# My RTX5080 GPU can't work with PyTorch

**URL:** <https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301>\
**Category:** Uncategorized\
**Created:** [February 28, 2025, 5:02pm UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301 "2025-02-28T17:02:11Z")\
**Posts on this page:** 20\
**Page:** 1

<div class="post-metadata">

**Author:** ![zp9450](https://discuss.pytorch.org/letter_avatar_proxy/v4/letter/z/7bcc69/32.png) [@zp9450](https://discuss.pytorch.org/u/zp9450)\
**Post date:** [February 28, 2025, 5:02pm UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/1 "2025-02-28T17:02:11Z")

</div>

\Local\Programs\Python\Python312\Lib\site-packages\torch\cuda\__init_\_.py:235: UserWarning:  
NVIDIA GeForce RTX 5080 with CUDA capability sm\_120 is not compatible with the current PyTorch installation.  
The current PyTorch install supports CUDA capabilities sm\_50 sm\_60 sm\_61 sm\_70 sm\_75 sm\_80 sm\_86 sm\_90.  
If you want to use the NVIDIA GeForce RTX 5080 GPU with PyTorch, please check the instructions at [Start Locally | PyTorch](https://pytorch.org/get-started/locally/)

and i tried [Start Locally | PyTorch](https://pytorch.org/get-started/locally/) and find nothing to help  
if there is an update suite my GPU  
please email me

---

<div class="post-metadata">

**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\
**Post date:** [March 1, 2025, 1:58am UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/2 "2025-03-01T01:58:46Z")

</div>

You need to install the nightly binaries with CUDA 12.8 (select nightly in the install matrix and replace the CUDA version in the URL with `cu128`). We’ll update the install matrix once all builds are ready.

---

<div class="post-metadata">

**Author:** ![ou17770103456](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ou17770103456/32/75907_2.png) [@ou17770103456](https://discuss.pytorch.org/u/ou17770103456)\
**Post date:** [March 1, 2025, 3:29am UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/3 "2025-03-01T03:29:27Z")

</div>

When will the new version of pytorch be released?

---

<div class="post-metadata">

**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\
**Post date:** [March 1, 2025, 1:03pm UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/4 "2025-03-01T13:03:29Z")

</div>

Nightly binaries are released every night. 2.7 is planned for April I believe.

---

<div class="post-metadata">

**Author:** ![zp9450](https://discuss.pytorch.org/letter_avatar_proxy/v4/letter/z/7bcc69/32.png) [@zp9450](https://discuss.pytorch.org/u/zp9450)\
**Post date:** [March 3, 2025, 6:11am UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/5 "2025-03-03T06:11:46Z")

</div>

pip3 install --pre torchvision torchaudio --index-url /whl/nightly/cu128  
Looking in indexes: /whl/nightly/cu128  
Collecting torchvision  
Using cached [https://download.pytorch.org/whl/nightly/cu128/torchvision-0.22.0.dev20250302%2Bcu128-cp312-cp312-win\_amd64.whl.metadata](https://download.pytorch.org/whl/nightly/cu128/torchvision-0.22.0.dev20250302%2Bcu128-cp312-cp312-win_amd64.whl.metadata) (6.3 kB)  
ERROR: Could not find a version that satisfies the requirement torchaudio (from versions: none)

---

<div class="post-metadata">

**Author:** ![nafe\_alhmdan](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/nafe_alhmdan/32/75984_2.png) [@nafe\_alhmdan](https://discuss.pytorch.org/u/nafe_alhmdan)\
**Post date:** [March 4, 2025, 7:31pm UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/6 "2025-03-04T19:31:18Z")

</div>

hi rtx5090 Error creating song: CUDA error: no kernel image is available for execution on the device Compile with TORCH\_USE\_CUDA\_DSA to enable device-side assertions.

---

<div class="post-metadata">

**Author:** ![zp9450](https://discuss.pytorch.org/letter_avatar_proxy/v4/letter/z/7bcc69/32.png) [@zp9450](https://discuss.pytorch.org/u/zp9450)\
**Post date:** [March 4, 2025, 9:32pm UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/7 "2025-03-04T21:32:35Z")

</div>

use preview version  
pip3 install --pre torch torchvision torchaudio --index-url [https://download.pytorch.org/whl/nightly/cu128](https://download.pytorch.org/whl/nightly/cu128)  
if you can‘t match the day package，check version in [https://download.pytorch.org/whl/torch/](https://download.pytorch.org/whl/torch/)

---

<div class="post-metadata">

**Author:** ![nafe\_alhmdan](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/nafe_alhmdan/32/75984_2.png) [@nafe\_alhmdan](https://discuss.pytorch.org/u/nafe_alhmdan)\
**Post date:** [March 4, 2025, 10:18pm UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/8 "2025-03-04T22:18:37Z")

</div>

how i can use preview version  
pip3 install --pre torch torchvision torchaudio --index-url

---

<div class="post-metadata">

**Author:** ![zp9450](https://discuss.pytorch.org/letter_avatar_proxy/v4/letter/z/7bcc69/32.png) [@zp9450](https://discuss.pytorch.org/u/zp9450)\
**Post date:** [March 4, 2025, 10:31pm UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/9 "2025-03-04T22:31:44Z")

</div>

“pip3 install --pre torch torchvision torchaudio --index-url [https://download.pytorch.org/whl/nightly/cu128”](https://download.pytorch.org/whl/nightly/cu128%E2%80%9D)  
put url end of pip3 cmd

---

<div class="post-metadata">

**Author:** ![nafe\_alhmdan](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/nafe_alhmdan/32/75984_2.png) [@nafe\_alhmdan](https://discuss.pytorch.org/u/nafe_alhmdan)\
**Post date:** [March 4, 2025, 11:04pm UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/10 "2025-03-04T23:04:31Z")

</div>

i can use anydesk fix my bc  
thank you

---

<div class="post-metadata">

**Author:** ![nafe\_alhmdan](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/nafe_alhmdan/32/75984_2.png) [@nafe\_alhmdan](https://discuss.pytorch.org/u/nafe_alhmdan)\
**Post date:** [March 5, 2025, 8:37pm UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/11 "2025-03-05T20:37:12Z")

</div>

hi  
ERROR: Cannot install torch, torchaudio==2.6.0 and torchvision==0.22.0.dev20250226+cu128 because these package versions have conflicting dependencies.

---

<div class="post-metadata">

**Author:** ![zp9450](https://discuss.pytorch.org/letter_avatar_proxy/v4/letter/z/7bcc69/32.png) [@zp9450](https://discuss.pytorch.org/u/zp9450)\
**Post date:** [March 6, 2025, 4:31am UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/12 "2025-03-06T04:31:27Z")

</div>

[https://download.pytorch.org/whl/nightly/torchvision/](https://download.pytorch.org/whl/nightly/torchvision/)  
find your version in this url  
i use torchvision-0.22.0.dev20250302+cu128-cp312-cp312-win\_amd64.whl  
choose your version with +cu128, cpu version，os type,cpu type

---

<div class="post-metadata">

**Author:** ![zp9450](https://discuss.pytorch.org/letter_avatar_proxy/v4/letter/z/7bcc69/32.png) [@zp9450](https://discuss.pytorch.org/u/zp9450)\
**Post date:** [March 6, 2025, 4:32am UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/13 "2025-03-06T04:32:17Z")

</div>

i can’t find torchaudio-2.6.0.dev20250302+cu128-cp312-cp312-win\_amd64.whl  
i need this version

---

<div class="post-metadata">

**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\
**Post date:** [March 6, 2025, 2:00pm UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/14 "2025-03-06T14:00:16Z")

</div>

> [@Pytorch support for sm120](https://discuss.pytorch.org/t/pytorch-support-for-sm120/216099/27):
>
> All audio builds fail on Windows due to: [[binary builds] torchaudio Windows cuda nightly builds are failing since 01.29.2025 · Issue #3877 · pytorch/audio · GitHub](https://github.com/pytorch/audio/issues/3877)

---

<div class="post-metadata">

**Author:** ![zp9450](https://discuss.pytorch.org/letter_avatar_proxy/v4/letter/z/7bcc69/32.png) [@zp9450](https://discuss.pytorch.org/u/zp9450)\
**Post date:** [March 6, 2025, 7:44pm UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/16 "2025-03-06T19:44:42Z")

</div>

i got torchvision-0.22.0.dev20250306+cu128-cp312-cp312-win\_amd64.whl  
but Requires-Dist: torch (==2.7.0.dev20250305+cu128)  
i edit 0305 to 0306 and install completed  
if there is a bug 0306 requires 0305,you should know

---

<div class="post-metadata">

**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\
**Post date:** [March 7, 2025, 8:27pm UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/17 "2025-03-07T20:27:18Z")

</div>

`torchaudio` binaries for Windows are unblocked now with a workaround.

---

<div class="post-metadata">

**Author:** ![nafe\_alhmdan](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/nafe_alhmdan/32/75984_2.png) [@nafe\_alhmdan](https://discuss.pytorch.org/u/nafe_alhmdan)\
**Post date:** [March 8, 2025, 11:57pm UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/18 "2025-03-08T23:57:13Z")

</div>

hi rtx 5090  
pip3 install --pre torch torchvision torchaudio --index-url [https://download.pytorch.org/whl/nightly/cu128](https://download.pytorch.org/whl/nightly/cu128) If you download the program  
replay  
Nvidia rvc  
it will work because I tried but it didn’t work

---

<div class="post-metadata">

**Author:** ![davidair](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/davidair/32/67785_2.png) [@davidair](https://discuss.pytorch.org/u/davidair)\
**Post date:** [April 7, 2025, 3:31am UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/19 "2025-04-07T03:31:30Z")

</div>

Just got an RTX5080, the following did the trick:

```auto
pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cu128

```

---

<div class="post-metadata">

**Author:** ![Aaargh\_Zombies](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/aaargh_zombies/32/74867_2.png) [@Aaargh\_Zombies](https://discuss.pytorch.org/u/Aaargh_Zombies)\
**Post date:** [May 25, 2025, 10:38am UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/20 "2025-05-25T10:38:01Z")

</div>

> pip install --pre torch torchvision torchaudio --index-url [https://download.pytorch.org/whl/nightly/cu128](https://download.pytorch.org/whl/nightly/cu128)

I’m not remotely technical and don’t really understand a lot of what’s being said on this thread so please bear with me.

I went to the Windows 10 command line and typed exactly what you said, and it came up with a long set of text that can mostly be boiled down to

> Requirement already satisfied

Followed by

> [notice] A new release of pip is available: 23.0.1 → 25.1.1

But I’m still getting the error message

> UserWarning:  
> NVIDIA GeForce RTX 5080 with CUDA capability sm\_120 is not compatible with the current PyTorch installation. The current PyTorch install supports CUDA capabilities sm\_50 sm\_60 sm\_61 sm\_70 sm\_75 sm\_80 sm\_86 sm\_90. If you want to use the NVIDIA GeForce RTX 5080 GPU with PyTorch, please check the instructions at [Start Locally | PyTorch](https://pytorch.org/get-started/locally/)

Is there anything else that I should be looking at?

I have just upgraded from a 30 series to a 50 series graphics card, and everything was running smoothly before that.

---

<div class="post-metadata">

**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\
**Post date:** [May 25, 2025, 2:13pm UTC](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301/21 "2025-05-25T14:13:59Z")

</div>

Uninstall your older PyTorch installation via `pip uninstall torch` before installing the new binary.

[Next page](https://discuss.pytorch.org/t/my-rtx5080-gpu-cant-work-with-pytorch/217301.md?page=2)
