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.
Fix
Public exploits 1
- Hunt-Benito/llama-factory-webui-rce-cve-2026-58116-trust-remote-code-model-path-injection PoC on GitHub
Each line links to the index that lists it, never to exploit code. Exploit-DB's "verified" and a Metasploit rank of good or better count as verified; a PoC-in-GitHub entry is a repository's claim.
Patch, vendor advisory and mitigation references on NVD 1
- https://gist.github.com/henrrrychau/08d76ec672f42136bbc1449c4f2973f8 Exploit, Mitigation, Third Party Advisory
Every line above is quoted from NVD, CISA, Microsoft, the vendor's own advisory or DISA; nothing is generated.
Change history 1
- First seenCVSS 9.3 · hiyougaNVD modified 2026-10-25 21:10
Affected (CPE criteria) 1
cpe:2.3:a:hiyouga:llama-factory:*:*:*:*:*:*:*:*
References 2
- 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