Description
PyTorch Lightning through 2.6.5, fixed in commit d710d68, contains a remote code execution vulnerability in the _load_state function that imports and executes attacker-controlled module names from checkpoint _instantiator hyperparameters. Attackers can craft malicious checkpoint files that bypass weights_only=True protections to execute arbitrary code when LightningModule.load_from_checkpoint is called.
Affected products
lightningai: pytorch lightning
References4
- github.com/Lightning-AI/pytorch-lightning/commit/d710d689510d50… Patch
- github.com/Lightning-AI/pytorch-lightning/pull/21832 Issue TrackingPatch
- www.vulncheck.com/advisories/pytorch-lightning-arbitrary-code-e… ExploitPatchThird Party Advisory
- github.com/Lightning-AI/pytorch-lightning/issues/21822 ExploitIssue TrackingMitigation
Metrics
- CVSS 3.1
-
7.8
HIGH
Score source: CNA
CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H - CVSS 4.0
-
8.4
HIGH
Score source: CNA
CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:A/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X - EPSS — exploit probability
- 1% percentile: 51 · as of 15 November 2026
- CWE
- CWE-470
Could this be exploited in your environment?
A patch closes the known hole. A penetration test shows what an attacker can still reach — with proof.