We use these services and cookies to improve your user experience. You may opt out if you wish, however, this may limit some features on this site.
Please see our statement on Data Privacy.
LLaMA-Factory is a tuning library for large language models. A remote code execution vulnerability was discovered in LLaMA-Factory versions up to and including 0.9.3 during the LLaMA-Factory training process. This vulnerability arises because the `vhead_file` is loaded without proper safeguards, allowing malicious attackers to execute arbitrary malicious code on the host system simply by passing a malicious `Checkpoint path` parameter through the `WebUI` interface. The attack is stealthy, as the victim remains unaware of the exploitation. The root cause is that the `vhead_file` argument is loaded without the secure parameter `weights_only=True`. Version 0.9.4 contains a fix for the issue.
Reserved 2025-06-24 | Published 2025-06-26 | Updated 2025-06-26 | Assigner GitHub_MCWE-94: Improper Control of Generation of Code ('Code Injection')
CWE-502: Deserialization of Untrusted Data
github.com/...actory/security/advisories/GHSA-xj56-p8mm-qmxj
github.com/...ommit/bb7bf51554d4ba8432333c35a5e3b52705955ede
drive.google.com/...sXfuvL4Tvbn_WJdjEOYXx4y/view?usp=sharing
Support options