CVE-2026-27893Disclosure(vllm / vllm)

LOWCVSS 8.8 · HIGH

Signal is active with 1 mentions in latest observed window

Immediate actions

  • Patch vllm vllm systems immediately

Recommended action window: Monitor and triage in normal cycle

NVD description

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.

0.5/ 10 priority

Sources & remediation

Weakness type (CWE)
CWE-693CWE-501

Priority

LOW

Exploitation

NONE

PoC

NONE

Patch

AVAILABLE

Momentum

STABLE

Are you affected?

If you run products in this scope, you should treat this CVE as relevant to your environment.

  • vllm

Threat summary

  • Patch or workaround signal is available
  • 4 mentions across 2 observed days
  • Momentum state: stable

What's happening

  • Patch or workaround mentioned in 1 signal
  • Technical details provided in 3 signals
  • Disclosure: 3 classified signals
  • Peaked 1d ago at 3 mentions (2026-03-27); latest day: 1
  • 4 total mentions across 2 days

Affected systems

Vendors
Products
vllm

Deep dive

Activity timeline4 mentions / 2d
01223Mentions · 2026-03-27: 3Mentions · 2026-03-28: 1Patch / Workaround · 2026-03-27: 1Technical Details · 2026-03-27: 2Technical Details · 2026-03-28: 103-2703-28
Signal classification2 categories
Disclosure
375.0%
Patch
125.0%
Referenced assets4 URLs
Classification over time
DateTotalLabels
2026-03-273
Disclosure2Patch1
2026-03-281
Disclosure1
Full discourse4 posts
  • The Hacker Wire@TheHackerWire
    Disclosure

    🟠 CVE-2026-27893 - High 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=Tr... https://www.thehackerwire.com/vulnerability/CVE-2026-27893/ https://t.co/OmrhsXvd4q

    Post summary

    The tweet announces high‑severity CVE‑2026‑27893 affecting vLLM, noting hardcoded parameters in certain versions and linking to a detailed vulnerability article.

    0000040
    163 followersView on X
  • CVE@CVEnew
    Disclosure

    CVE-2026-27893 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 hardc… https://www.cve.org/CVERecord?id=CVE-2026-27893

    Post summary

    The text announces the discovery of CVE‑2026‑27893 in vLLM, noting affected versions, but it provides no PoC, exploit code, patch, or detailed technical specifics.

    00000118
    56.9K followersView on X
  • CVEarity@CVEarity
    Disclosure

    ⚡ New CVE Alert: CVE-2026-27893 📊 Severity: 8.8 🚨 Risk Level: High 🧩 Affects: Multiple / Unspecified Products Reference: https://nvd.nist.gov/vuln/detail/CVE-2026-27893 #CVE-2026-27893 #CVE #High #CyberSecurity #InfoSec https://t.co/mw3cfQP2Am

    Post summary

    A new high‑severity vulnerability, CVE‑2026‑27893, has been reported with a CVSS score of 8.8 and affects multiple unspecified products, with reference to the NVD description.

    0000028
    123 followersView on X
  • CVEFind.com@CveFindCom
    Patch

    [CVE-2026-27893: HIGH] vLLM versions 0.10.1 to 0.18.0 had a security flaw allowing remote code execution despite user settings. Update to version 0.18.0 to fix this issue. #cybersecurity#cve,CVE-2026-27893,#cybersecurity https://cvefind.com/CVE-2026-27893

    Post summary

    A critical RCE vulnerability is disclosed in vLLM 0.10.1‑0.18.0, and users are urged to upgrade to v0.18.0 to mitigate the issue.

    0000063
    617 followersView on X
CPE platform detail1 entries

1 of 1 entries

PartVendorProductVersionTarget SWTarget HW
Appvllmvllm---

Explore more