LiveActive security incident?Get immediate response
CVE Record

CVE-2026-27893: vLLM's hardcoded trust_remote_code=True in NemotronVL and KimiK25 bypasses user security opt-out

vLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1 and prior to version 0.18.0, two model implementation files hardcode `trust_remote_code=True` when loading sub-components, bypassing the user's explicit `--trust-remote-code=False` security opt-out. This enables remote code execution via malicious model repositories even when the user has explicitly disabled remote code trust. Version 0.18.0 patches the issue.

HighCVSS 8.8Not KEV-listedUpdated
Discuss exposureBack to CVEs
Glexia's TakeAutomated analysishigh

Security readout for executives and security teams

Plain-English summary

vLLM versions 0.10.1 through before 0.18.0 can run code from a malicious model repository even when an operator explicitly disabled remote-code trust. This turns a safety setting into a false assurance for affected model-loading paths.

Executive priority

Treat as high priority for teams running vLLM in production or shared research platforms. The main business risk is code execution from a model source despite a configured safety opt-out, which can undermine trust in AI platform controls.

Technical view

NemotronVL and KimiK25 implementation files hardcoded `trust_remote_code=True` when loading sub-components, bypassing `--trust-remote-code=False`. A malicious model repository could execute code during loading. The advisory states vLLM 0.18.0 fixes the issue.

Likely exposure

Exposure is most likely in AI inference environments running vLLM >=0.10.1 and <0.18.0 that load model repositories, especially using affected NemotronVL or KimiK25 code paths. Red Hat advisories are referenced, but product applicability should be confirmed per erratum.

Exploitation context

The source bundle does not show CISA KEV listing or cited evidence of active exploitation. The CVSS vector indicates network reachability, low complexity, no privileges, and required user interaction, consistent with loading a malicious model repository.

Researcher notes

Primary evidence is the GitHub advisory, fix PR, and commit. CWE mappings are CWE-501 and CWE-693. The stated affected range is >=0.10.1 and <0.18.0. Do not assume broader products beyond vLLM or named vendor errata without checking those advisories.

Mitigation direction

  • Upgrade vLLM to version 0.18.0 or later.
  • Check Red Hat errata for packaged product fixes and applicability.
  • Restrict model loading to trusted, approved repositories.
  • Review pipelines that rely on `--trust-remote-code=False` as a control.
  • Monitor vendor advisories for additional mitigations or backports.

Validation and detection

  • Inventory vLLM versions in containers, notebooks, inference services, and build files.
  • Identify deployments using NemotronVL, KimiK25, or related model-loading paths.
  • Confirm affected systems are upgraded to vLLM 0.18.0 or fixed vendor builds.
  • Verify model-source allowlists and approval controls are enforced.
  • Review logs for unexpected model repository loads before remediation.
Prepared
Confidence
high
Sources
10

Public sources used

Generated from the cited source records. This long-tail analysis has not been individually reviewed by a named human.

Potential ATT&CK relevance

Conservative CVE-to-ATT&CK context

These mappings and lookup hints may be relevant to the vulnerability behavior, CWE, affected product, or exposure path. Glexia-inferred context is not an official MITRE, ATT&CK, CWE, or CVE Program mapping.

ATT&CK lookup starting points

Use these exact CWE pages and searches to review the Glexia ATT&CK library from this CVE's weakness and description context.

cwe · low confidence lookup

CWE-501: Exact CWE lookup

Use the exact CWE identifier as the starting point before reviewing related ATT&CK behavior. Open the exact CWE lookup page first, then review the ATT&CK searches from that MITRE weakness context. This is a Glexia lookup hint, not an official ATT&CK mapping.

Open ATT&CK lookup
cwe · low confidence lookup

CWE-693: Exact CWE lookup

Use the exact CWE identifier as the starting point before reviewing related ATT&CK behavior. Open the exact CWE lookup page first, then review the ATT&CK searches from that MITRE weakness context. This is a Glexia lookup hint, not an official ATT&CK mapping.

Open ATT&CK lookup
description · low confidence lookup

Execution behavior lookup

The CVE wording references code or command execution, so execution technique review may help defensive triage. This is a Glexia inferred lookup path, not an official MITRE, ATT&CK, or CVE Program mapping.

Open ATT&CK lookup
cve · low confidence lookup

CVE-2026-27893 mapping review

Open the CVE-to-ATT&CK bridge for reviewed, inferred, or future official mappings tied to this CVE.

Open ATT&CK lookup
Vulnerability profileCVE Program record
Severity
High
CVSS
8.8 (3.1)
Known Exploited
No
Published

Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Official CVE source material

CNA and ADP enrichment extracted from CVE v5

These fields come from the CVE record and ADP containers, not from Glexia's Take. They preserve time-varying source decisions such as CISA SSVC, KEV status, CVSS metrics, and provider references.

2CVSS vectors
5Timeline events
2ADP providers
17Source links

SSVC decision data

CISA-ADPCISA Coordinator
Timestamp
Version
2.0.3
Exploitation: noneAutomatable: noTechnical Impact: total

CVSS vector scores

2 official scores

We collect every scored CVSS vector available in the official CNA and ADP containers. When more than one version is present, the table keeps the source vectors side by side instead of collapsing them into the highest score.

ScoreVersionSeverityVectorExploitImpactSource
8.8CVSS 3.1HighCVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H2.85.9GitHub_M
8.8CVSS 3.1HighCVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H2.85.9redhat-SADP

Vulnerability scoring details

Base CVSS 3.1 score

8.8High
CVSS 3.1 vector shape for CVE-2026-27893Attack VectorAttack ComplexityPrivileges RequiredUser InteractionScopeConfidentiality ImpactIntegrity ImpactAvailability Impact

Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Attack Vector
NetworkAdjacentLocalPhysical
Attack Complexity
LowHigh
Privileges Required
NoneLowHigh
User Interaction
NoneRequired
Scope
ChangedUnchanged
Confidentiality Impact
HighLowNone
Integrity Impact
HighLowNone
Availability Impact
HighLowNone

Vulnerability timeline

Timeline events are normalized from CVE metadata, CNA source timelines, ADP timelines, and KEV metadata when present.

  1. CVE reservedCVE Program

    The CVE ID was reserved by the assigning CNA.

  2. ADP timelineredhat-SADP

    Made public.

  3. CVE publishedCVE Program

    The CVE record was published.

  4. ADP timelineredhat-SADP

    Reported to Red Hat.

  5. CVE updatedCVE Program

    The CVE record metadata indicates this as the latest update time.

ADP provider summaries

CISA-ADPCISA ADP Vulnrichment
other:ssvc
redhat-SADPvllm: vLLM: Remote code execution due to hardcoded trust_remote_code setting
other:Red Hat severity ratingcvssV3_1
  • 2026-03-27T00:01:43.935Z: Reported to Red Hat.
  • 2026-03-26T23:56:53.579Z: Made public.

Source materials

Affected products

Products and packages named in the record

VendorProductVersion / packageStatus
vllm-projectvllm>= 0.10.1, < 0.18.0Listed
Weakness

CWE details

CWE-501CWE-693

CWE links open Glexia weakness intelligence pages with official CWE context, developer remediation guidance, and related CVE mappings.