Vulnerability details
CVE-2026-4944
vllm-project vllm-project/vllm
Hardcoded trust_remote_code=True in vllm-project/vllm Bypasses User Security Control
CVSS 3.0
8.8High
Official 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.
CVSS scores
| Score | Version | Source | Vector |
|---|---|---|---|
| 8.8High | 3.0 | CVEProject cvelistV5 (CNA) (@huntr_ai) | CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H |
Affected products
| Vendor | Product | Versions | Source |
|---|---|---|---|
| vllm-project | vllm-project/vllm | v ≤ latest | CVEProject cvelistV5 (CNA) |
References and advisories
1
Kolun on Telegram @kolunsec
Exploited CVEs, key news and a daily digest
Alerts for this product's next CVEsFollow the product to get new vulnerabilities on Telegram.
Details
- Published
- 2026-05-28
- Last updated
- 2026-05-28
- State
- Published
- Assigner (CNA)
- @huntr_ai
Exploit prediction (EPSS)
0.74%
Probability of exploitation in 30 days
55.1%
Percentile (among all CVEs)
SSVC assessment (CISA)
- Automatable
- No
- Exploitation
- Proof of concept
- Technical impact
- Total
Weaknesses (CWE)
- CWE-22Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal')