A vulnerability in the LightGlue model loading path of huggingface/transformers version 5.2.0 allows an attacker-controlled model repository to execute arbitrary code during model initialization. The issue arises because the `trust_remote_code` parameter, intended to prevent remote code execution, is overridden by untrusted serialized configuration data in a nested code path. Specifically, when loading a LightGlue model using `AutoModel.from_pretrained()` with `trust_remote_code=False`, the `LightGlueConfig` reads the `trust_remote_code` value from the untrusted `config.json` file and propagates it into nested `AutoConfig.from_pretrained()` calls. This results in the execution of attacker-provided Python modules, even when the victim explicitly disables remote code execution. The vulnerability poses a high risk for environments such as API inference servers, research notebooks, CI/CD pipelines, and model evaluation workers, potentially leading to credential theft, lateral movement, or persistence/backdoor deployment.
| Vendor | Product | Version(s) | CPE |
|---|---|---|---|
5.2.0CPE matchmatch criteria | cpe:2.3:a:huggingface:transformers:5.2.0:*:*:*:*:*:*:* |
CVSS version used by this source: 3.0
CVSS:3.0/AV:N/AC:H/PR:N/UI:R/S:C/C:H/I:H/A:N
The average CVE in this peer group has 0.0 Twitter, 0.1 Reddit, 0.8 Bluesky, 0.5 Mastodon, and 1.2 GitHub mentions.
No media coverage found for this CVE.
The average CVE in this peer group has 0.3 InfoSec Media, 0.0 Vendor Blog, and 0.1 Security Researcher mentions.