Hi, I install diffuser in a server following this guide (Installation) and use the following command to install it (pip install diffusers[“torch”] transformers). However, the command automatically installs the latest version of PyTorch (PyTorch 2.1.0) and the latest version of packages that can run only on CUDA 12 and higher CUDA versions. However, my local server’s CUDA version is only 10.2 so these packages with latest version can’t run on the server. How can I let the installed packages compatible with my local CUDA 10.2?
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Conflict in pytorch CUDA version and space CUDA version | 0 | 365 | May 25, 2024 | |
| Device Type Error With Diffusers Pipeline TAT | 2 | 1301 | August 19, 2023 | |
| ERROR: Could not find a version that satisfies the requirement torch==1.12.1+cu102 | 1 | 3781 | November 5, 2023 | |
| How do I fix the "RuntimeError: CUDA error: CUDA driver version is insufficient for CUDA runtime version CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect." error? | 12 | 89212 | July 13, 2023 | |
| Diffusers course introduction-to-diffusers setup fails | 2 | 20 | May 20, 2025 |