| Package | Affected versions | Patched versions |
|---|---|---|
| torch | < 2.6.0 | 2.6.0 |
| pytorch | < 2.6.0 | 2.6.0 |
I found a Remote Command Execution (RCE) vulnerability in PyTorch. When loading model using torch.load with weights_only=True, it can still achieve RCE.
https://github.com/pytorch/pytorch/security
As you can see, the PyTorch official documentation considers using torch.load() with weights_only=True to be safe.
Since everyone knows that weights_only=False is unsafe, so they will use the weights_only=True to mitigate the seucirty issue.
But now, I just proved that even if you use weights_only=True, it can still achieve RCE.
This vulnerability was found by Ji'an Zhou.
Running one of these products? Scan your site to see whether it exposes a vulnerable version.
Scan my site free