CVE-2026-4944
Hardcoded trust_remote_code=True in vllm-project/vllm Bypasses User Security Control
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.
INFO
Published Date :
May 28, 2026, 7:16 p.m.
Last Modified :
May 29, 2026, 3:39 p.m.
Remotely Exploit :
Yes !
Source :
[email protected]
CVSS Scores
| Score | Version | Severity | Vector | Exploitability Score | Impact Score | Source |
|---|---|---|---|---|---|---|
| CVSS 3.0 | HIGH | [email protected] |
Solution
- Remove hardcoded trust_remote_code=True.
- Validate all model sources.
- Apply code changes.
References to Advisories, Solutions, and Tools
Here, you will find a curated list of external links that provide in-depth
information, practical solutions, and valuable tools related to
CVE-2026-4944.
| URL | Resource |
|---|---|
| https://huntr.com/bounties/97f706f7-a852-49b2-a4eb-76811e611daf | |
| https://huntr.com/bounties/97f706f7-a852-49b2-a4eb-76811e611daf |
CWE - Common Weakness Enumeration
While CVE identifies
specific instances of vulnerabilities, CWE categorizes the common flaws or
weaknesses that can lead to vulnerabilities. CVE-2026-4944 is
associated with the following CWEs:
Common Attack Pattern Enumeration and Classification (CAPEC)
Common Attack Pattern Enumeration and Classification
(CAPEC)
stores attack patterns, which are descriptions of the common attributes and
approaches employed by adversaries to exploit the CVE-2026-4944
weaknesses.
We scan GitHub repositories to detect new proof-of-concept exploits. Following list is a collection of public exploits and proof-of-concepts, which have been published on GitHub (sorted by the most recently updated).
Results are limited to the first 15 repositories due to potential performance issues.
The following list is the news that have been mention
CVE-2026-4944 vulnerability anywhere in the article.
The following table lists the changes that have been made to the
CVE-2026-4944 vulnerability over time.
Vulnerability history details can be useful for understanding the evolution of a vulnerability, and for identifying the most recent changes that may impact the vulnerability's severity, exploitability, or other characteristics.
-
CVE Modified by 134c704f-9b21-4f2e-91b3-4a467353bcc0
May. 28, 2026
Action Type Old Value New Value Added Reference https://huntr.com/bounties/97f706f7-a852-49b2-a4eb-76811e611daf -
New CVE Received by [email protected]
May. 28, 2026
Action Type Old Value New Value Added 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. Added CVSS V3 AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H Added CWE CWE-22 Added Reference https://huntr.com/bounties/97f706f7-a852-49b2-a4eb-76811e611daf