This version of PyTorch is not compatible with RTX 5000 GPUs

[freemansoft]

AI Workbench version: "latest" ? don't see an about panel. installed via remote command line today.
Ubuntu: 24.04
Video Card: 5060 TI

Saw this same problem in one of the NIMs, but AI Workbench team owns this project and not the NIM club.

-- MODELS: Loading Model stabilityai/stable-diffusion-xl-base-1.0 ---

Fetching 19 files:   0%|          | 0/19 [00:00<?, ?it/s]
Fetching 19 files:  11%|█         | 2/19 [00:00<00:05,  2.96it/s]
Fetching 19 files:  21%|██        | 4/19 [00:59<04:20, 17.38s/it]
Fetching 19 files:  32%|███▏      | 6/19 [01:02<02:12, 10.19s/it]
Fetching 19 files:  84%|████████▍ | 16/19 [01:10<00:09,  3.01s/it]
Fetching 19 files: 100%|██████████| 19/19 [01:10<00:00,  3.72s/it]

Loading pipeline components...:   0%|          | 0/7 [00:00<?, ?it/s]
Loading pipeline components...:  14%|█▍        | 1/7 [00:00<00:01,  5.00it/s]
Loading pipeline components...:  29%|██▊       | 2/7 [00:00<00:00,  5.01it/s]
Loading pipeline components...:  86%|████████▌ | 6/7 [00:00<00:00, 13.73it/s]
Loading pipeline components...: 100%|██████████| 7/7 [00:00<00:00,  8.84it/s]
/home/workbench/.local/lib/python3.10/site-packages/torch/cuda/__init__.py:235: UserWarning: 

NVIDIA GeForce RTX 5060 Ti 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_70 sm_75 sm_80 sm_86 sm_90.
If you want to use the NVIDIA GeForce RTX 5060 Ti GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/
  warnings.warn(
INFO:httpx:HTTP Request: GET https://checkip.amazonaws.com/ "HTTP/1.1 200 "
INFO:httpx:HTTP Request: GET https://api.gradio.app/pkg-version "HTTP/1.1 200 OK"
INFO:httpx:HTTP Request: GET http://localhost:8080/startup-events "HTTP/1.1 200 OK"
INFO:httpx:HTTP Request: HEAD http://localhost:8080/ "HTTP/1.1 200 OK"
INFO:matplotlib.font_manager:generated new fontManager
--- MODELS: Configuring Pipe ---
--- MODELS: Model is ready for inference ---
http://localhost:8000
IMPORTANT: You are using gradio version 4.35.0, however version 4.44.1 is available, please upgrade.
--------
Running on local URL:  http://0.0.0.0:8080
To create a public link, set `share=True` in `launch()`.

[ekzotech]

I was able to fix this warning. Not sure if everything works as expected.

1. Add to the apt.txt file python3-pip package
2. Bump versions for torch in the requirements.txt file: torch>=2.9.0.dev torchvision>=0.24.0.dev torchaudio>=2.9.0.dev. Also you have to bump nvidia versions to handle CUDA 12.8: nvidia-cublas-cu12>=12.4.5.8 nvidia-cuda-cupti-cu12>=12.8.0 nvidia-cuda-nvrtc-cu12>=12.8.0 nvidia-cuda-runtime-cu12>=12.8.0 nvidia-cufft-cu12>=11.2.1.3 nvidia-curand-cu12>=10.3.5.147 nvidia-cusolver-cu12>=11.6.1.9 nvidia-cusparse-cu12>=12.3.1.170 nvidia-cusparselt-cu12>=0.6.2 nvidia-dali-cuda120>=1.29.0 nvidia-nvjitlink-cu12>=12.8.0 nvidia-nvtx-cu12>=12.8.0 nvidia-pyindex>=1.0.9
   UPD: you need to change sympy version from 1.13.1 to at least 1.13.3
3. Add line pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cu128 to the preBuild.bash file to allow pip to use nightly builds of torch

After that you should be able to rebuild the containers and get into Jupyter Notebook without warnings.