Cyber Resilience

CVE-2025-66448

RCE in Vllm ≤ 0.11.1

Published
01 December 2025
Modified
03 December 2025
Patch / advisory
CVSS Score v3.1 7.1
Click a component to see what it means
Raw vectorCVSS:3.1/AV:N/AC:H/PR:L/UI:R/S:U/C:H/I:H/A:H
EPSS Score 0.0067 49th percentile
Risk Priority 53 floored blend · peak EPSS

Summary

CVE-2025-66448 is a high-severity Code Injection (CWE-94) vulnerability in Vllm Vllm. Its CVSS base score is 7.1 (High).

Operationally, exploitation aligns with the MITRE ATT&CK technique Command and Scripting Interpreter (T1059); ranked at the 49th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.

This vulnerability is AI-related — categorised as NLP and Transformers; in the Supply Chain and Deployment risk domain.

The strongest mitigations our analysis identified map to SA-11 (Developer Testing and Evaluation) and SI-10 (Information Input Validation) — see the control section below for these in your framework.

Deeper analysis AI-assisted summary

Synthesised by an AI model from the NVD description and linked references — a reading aid, not an authoritative source.

CVE-2025-66448 is a remote code execution vulnerability in vLLM, an inference and serving engine for large language models, affecting versions prior to 0.11.1. The issue resides in the Nemotron_Nano_VL_Config class, where loading a model configuration containing an auto_map entry triggers resolution via get_class_from_dynamic_module, which fetches and instantiates Python code from a remote repository specified in the auto_map string. This execution occurs even when the caller explicitly sets trust_remote_code=False in vllm.transformers_utils.config.get_config, bypassing intended security controls.

An attacker can exploit this by publishing a seemingly benign frontend repository with a config.json file that includes an auto_map pointing to a separate malicious backend repository. A victim loading the frontend model config will silently fetch and execute the backend's arbitrary Python code on their host. Per the CVSS v3.1 score of 7.1 (AV:N/AC:H/PR:L/UI:R/S:U/C:H/I:H/A:H), exploitation requires network access, high attack complexity, low privileges, and user interaction, but yields high impacts on confidentiality, integrity, and availability, classified under CWE-94 (code injection).

The vulnerability is addressed in vLLM version 0.11.1. Official mitigation details are available in the project's security advisory at GHSA-8fr4-5q9j-m8gm, the fixing pull request at github.com/vllm-project/vllm/pull/28126, and the commit ffb08379d8870a1a81ba82b72797f196838d0c86, which practitioners should review for patch implementation guidance.

This flaw highlights risks in AI/ML inference engines handling untrusted model configurations from remote sources, with no reported real-world exploitation as of the CVE publication on 2025-12-01.

OWASP Top 10 for Web (2025)

EU & UK References

Vulnerability Data

vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.11.1, vllm has a critical remote code execution vector in a config class named Nemotron_Nano_VL_Config. When vllm loads a model config that contains an auto_map entry,…

more

the config class resolves that mapping with get_class_from_dynamic_module(...) and immediately instantiates the returned class. This fetches and executes Python from the remote repository referenced in the auto_map string. Crucially, this happens even when the caller explicitly sets trust_remote_code=False in vllm.transformers_utils.config.get_config. In practice, an attacker can publish a benign-looking frontend repo whose config.json points via auto_map to a separate malicious backend repo; loading the frontend will silently run the backend’s code on the victim host. This vulnerability is fixed in 0.11.1.

CWE(s)

AI Security AnalysisAI

AI Category
NLP and Transformers
Risk Domain
Supply Chain and Deployment
OWASP Top 10 for LLMs 2025
None mapped
Classification Reason
Matched keywords: llms, vllm

Related Threats

MITRE ATT&CK Enterprise Techniques

T1059 Command and Scripting Interpreter Execution
Adversaries may abuse command and script interpreters to execute commands, scripts, or binaries.
T1059.001 PowerShell Execution
Adversaries may abuse PowerShell commands and scripts for execution.
T1059.002 AppleScript Execution
Adversaries may abuse AppleScript for execution.
T1059.004 Unix Shell Execution
Adversaries may abuse Unix shell commands and scripts for execution.
T1059.005 Visual Basic Execution
Adversaries may abuse Visual Basic (VB) for execution.
T1059.006 Python Execution
Adversaries may abuse Python commands and scripts for execution.
Derived from this CVE’s CWE(s) via the direct CWE→ATT&CK cross-walk.

CVEs Like This One

CVE-2026-22807Same product: Vllm Vllm
CVE-2026-41523Same product: Vllm Vllm
CVE-2026-34760Same product: Vllm Vllm
CVE-2025-48944Same product: Vllm Vllm
CVE-2026-44222Same product: Vllm Vllm
CVE-2026-44223Same product: Vllm Vllm
CVE-2025-62372Same product: Vllm Vllm
CVE-2025-62164Same product: Vllm Vllm
CVE-2025-30165Same product: Vllm Vllm
CVE-2025-32444Same product: Vllm Vllm

Affected Assets

vllm
vllm
≤ 0.11.1

Mitigating Controls

Control response

Prevent
Stop it (NIST 800-53)

Detect
Catch it (NIST detect / respond)

Harden
Shrink the surface (DISA STIG)

Validate
Prove the fix (OWASP ASVS)
  • V1.3.1

Mitigating Controls (NIST 800-53 r5) AI

Developer testing and evaluation finds code paths that accept and execute externally influenced strings.

Input validation directly stops untrusted data from being used to construct executable code without neutralization.

Least privilege limits the damage an injected code fragment can perform once executed.

Requiring documented secure development standards and tools enforces use of safe code-generation APIs and escaping.

Mitigating Controls (NIST CSF 2.0) AI

Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.

PR.PS-06 mostly match
prevents

PR.PS-06's SDLC practices directly target injection flaws via secure coding and testing (mostly), yet as a single broad outcome it leaves many code-generation specifics unaddressed (partial).

PR.DS-10 none match
prevents

PR.DS-10 protects runtime data confidentiality/integrity but has no bearing on neutralizing externally influenced input during code generation, so neither direction shows any preventive effect.

Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI

Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.

prevents

Banning unapproved code samples and unauthenticated web services, combined with secure-coding standards and SAST, prevents the dynamic generation or inclusion of attacker-supplied code.

none

Controls that restrict unauthorized or malicious code from being introduced via external networks or removable media limit opportunities for an attacker to inject and execute arbitrary code.

References