LLaMA-Factory through 0.9.5 contains a remote code execution vulnerability that allows attackers with WebUI access to execute arbitrary Python code by supplying a malicious model path in the Chat or Training interfaces. The application passes user-supplied model path input unvalidated into AutoTokenizer.from_pretrained() and AutoModel.from_pretrained() with a hardcoded trust_remote_code=True parameter, causing the Hugging Face transformers library to fetch and execute arbitrary code from a remote or local model repository with the privileges of the server process.
References
| Link | Resource |
|---|---|
| https://gist.github.com/henrrrychau/08d76ec672f42136bbc1449c4f2973f8 | Exploit Mitigation Third Party Advisory |
| https://www.vulncheck.com/advisories/llama-factory-remote-code-execution-via-webui-model-path | Third Party Advisory |
Configurations
History
No history.
Information
Published : 2026-06-30 13:19
Updated : 2026-07-14 22:17
NVD link : CVE-2026-58116
Mitre link : CVE-2026-58116
CVE.ORG link : CVE-2026-58116
JSON object : View
Products Affected
hiyouga
- llama-factory
