Home

Description

vllm-project/vllm version 0.14.1 contains a vulnerability where the `trust_remote_code=True` parameter is hardcoded in two model implementation files (`vllm/model_executor/models/nemotron_vl.py` and `vllm/model_executor/models/kimi_k25.py`). This bypasses the user's explicit `--trust-remote-code=False` setting, enabling remote code execution via malicious HuggingFace model repositories. This issue is an incomplete fix for CVE-2025-66448 and CVE-2026-22807, as it affects separate code paths in model implementation files. Deployments loading NemotronVL or KimiK25 models are particularly impacted.

PUBLISHED Reserved 2026-03-26 | Published 2026-05-28 | Updated 2026-05-28 | Assigner @huntr_ai




HIGH: 8.8CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Problem types

CWE-22 Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal')

Product status

Any version
affected

References

huntr.com/bounties/97f706f7-a852-49b2-a4eb-76811e611daf exploit

huntr.com/bounties/97f706f7-a852-49b2-a4eb-76811e611daf

cve.org (CVE-2026-4944)

nvd.nist.gov (CVE-2026-4944)

Download JSON