Cyber Resilience

CVE-2025-32444

RCE in Vllm 0.6.5 – 0.8.5

Published
30 April 2025
Modified
17 June 2026
Patch / advisory
CVSS Score v3.1 10.0
Click a component to see what it means
Raw vectorCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H
EPSS Score 0.017 75th percentile
Risk Priority 82 floored blend · peak EPSS

Summary

CVE-2025-32444 is a critical-severity Deserialization of Untrusted Data (CWE-502) vulnerability in Vllm Vllm. Its CVSS base score is 10.0 (Critical).

Operationally, exploitation aligns with the MITRE ATT&CK technique Exploit Public-Facing Application (T1190); ranked in the top 25% of CVEs by exploit likelihood; it is not currently listed in the CISA KEV catalog; a public proof-of-concept is referenced.

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.

vLLM is a high-throughput inference and serving engine for large language models. CVE-2025-32444 affects versions 0.6.5 through 0.8.4 that enable the optional mooncake integration. The flaw stems from the use of Python pickle serialization over unauthenticated ZeroMQ sockets that bind to all network interfaces, allowing deserialization of untrusted data (CWE-502) and resulting in a CVSS 10.0 remote code execution vector.

An attacker reachable to the ZeroMQ endpoints can send a malicious pickle payload and execute arbitrary code on the vLLM host without authentication or user interaction. Only deployments that activate mooncake are exposed; instances that do not use this integration remain unaffected.

The project has released version 0.8.5, which disables the vulnerable code path. Remediation guidance and the fixing commit are documented in the GitHub Security Advisories GHSA-hj4w-hm2g-p6w5 and GHSA-x3m8-f7g5-qhm7, along with the associated pull request that removes the insecure socket configuration.

The EPSS score rose from a low baseline to a recorded peak of 0.0776, indicating emerging exploitation interest after disclosure and suggesting the issue merits renewed monitoring in LLM-serving environments.

OWASP Top 10 for Web (2025)

EU & UK References

Vulnerability Data

vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Versions starting from 0.6.5 and prior to 0.8.5, having vLLM integration with mooncake, are vulnerable to remote code execution due to using pickle based serialization over unsecured ZeroMQ…

more

sockets. The vulnerable sockets were set to listen on all network interfaces, increasing the likelihood that an attacker is able to reach the vulnerable ZeroMQ sockets to carry out an attack. vLLM instances that do not make use of the mooncake integration are not vulnerable. This issue has been patched in version 0.8.5.

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

T1190 Exploit Public-Facing Application Initial Access
Adversaries may attempt to exploit a weakness in an Internet-facing host or system to initially access a network.
T1203 Exploitation for Client Execution Execution
Adversaries may exploit software vulnerabilities in client applications to execute code.
T1210 Exploitation of Remote Services Lateral Movement
Adversaries may exploit remote services to gain unauthorized access to internal systems once inside of a network.
Derived from this CVE’s CWE(s) via the direct CWE→ATT&CK cross-walk.

CVEs Like This One

CVE-2025-30165Same product: Vllm Vllm
CVE-2025-29783Same product: Vllm Vllm
CVE-2024-11041Same product: Vllm Vllm
CVE-2025-47277Same product: Vllm Vllm
CVE-2025-24357Same product: Vllm Vllm
CVE-2025-62164Same 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

Affected Assets

vllm
vllm
0.6.5 — 0.8.5

Mitigating Controls

Mitigating Controls (NIST 800-53 r5) AI

Developer testing and evaluation can uncover deserialization flaws before deployment.

Input validation directly stops deserialization of untrusted data by ensuring inputs are valid before processing.

Engineering principles such as safe deserialization and input sanitization structurally prevent the weakness from being introduced.

Integrity verification tools can detect malformed or tampered serialized data after the fact.

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-02 none match
prevents

PR.PS-02 addresses only post-deployment updates/patching and cannot prevent introduction of unsafe deserialization code, yet it can remediate some instances when the flaw exists in outdated libraries or components.

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.

finds

Security testing includes validation of deserialization routines and the use of untrusted data, reducing the likelihood that unsafe object reconstruction will be deployed.

prevents

Requiring vetted libraries, regular updates and SAST before release reduces the likelihood that deserialization logic will accept and act on attacker-controlled serialized objects.

finds

Regular scanning of third-party libraries and timely patching reduce the likelihood that unsafe deserialization vulnerabilities remain active.

none

Mandatory malware scanning of data received over networks or storage media intercepts malicious serialized payloads before they are deserialized by the target application.

References