Skip to main content
VulnerabilityPublished Updated

CVE-2026-58116: hiyouga LlamaFactory · Untrusted Functionality Inclusion

Affects hiyouga LlamaFactorySource CVE Database V5

0
Compare

Not known to be exploited · CVSS 9.3 Critical.

Data as of 12 Nov 2026, 23:24 UTC

CVSS 4.0 base score
9.3Critical
How severe the flaw is on the 0–10 CVSS scale.
EPSS
0.9% chance of exploitation in 30 days – higher than 60% of CVEsExploit Prediction Scoring System: the likelihood attackers will use this flaw soon.
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N

Exploitation evidence

  1. Disclosed: evidence found.

    CVE published 30 Jun 2026

  2. Detectable: no evidence on record.

    No public nuclei template indexed

  3. Weaponized: evidence found.

    Public exploit code: GitHub proofs of concept

  4. Exploited in the wild: no evidence on record.

    Not in CISA KEV and no source reports exploitation

  5. Observed by OffSeq Mirage: no evidence on record.

    No attack traffic recorded by OffSeq Mirage sensors

A missing step is not evidence of safety. Sources: cve.org (opens in a new tab), GitHub proofs of concept, the Metasploit, Exploit-DB and nuclei-templates index as of 13 Nov 2026.

What to do

Affected software

ProductAffectedFixed in
hiyougaLlamaFactoryAffected
  • <=0.9.5
Fixed in–
GitHubLlamaFactorypkg:github/LlamaFactoryOther threatsAffected
  • <=0.9.5
Fixed in–

Check your own hosts for these packages with threat-finder (opens in a new tab), our free open-source scanner.

AI-suggested

Until a fix is available, restrict or disable WebUI access to trusted users only. Avoid supplying untrusted model paths to the application. Monitor for updates from the vendor regarding patches or official mitigations.

Description

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

Community reviews

Loading reviews…