Cyber Resilience

CVE-2026-48746

Vllm 0.3.0 – 0.22.0

Published
22 June 2026
Modified
11 August 2026
CVSS Score v3.1 9.1
Click a component to see what it means
Raw vectorCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:H
EPSS Score 0.012 64th percentile
Risk Priority 69 floored blend · peak EPSS

Summary

CVE-2026-48746 is a critical-severity HTTP Request/Response Smuggling (CWE-444) vulnerability in Vllm Vllm. Its CVSS base score is 9.1 (Critical).

Operationally, exploitation aligns with the MITRE ATT&CK technique Exploit Public-Facing Application (T1190); ranked in the top 36% of CVEs by exploit likelihood; it is not currently listed in the CISA KEV catalog.

This vulnerability is AI-related — categorised as APIs and Models.

The strongest mitigations our analysis identified map to AC-4 (Information Flow Enforcement) and AC-16 (Security and Privacy Attributes) — see the control section below for these in your framework.

OWASP Top 10 for Web (2025)

A06:2025 Insecure Design

EU & UK References

Vulnerability Data

vLLM is an inference and serving engine for large language models (LLMs). From 0.3.0 until 0.22.0, a vulnerability in ASGI web servers and starlette's trust on those web servers enables an authentication bypass of the OpenAI API AuthenticationMiddleware. It allows…

more

to use the API without providing the configured VLLM_API_KEY or --api-key. This vulnerability is fixed in 0.22.0.

CWE(s)
CWE-444CWE-501

AI Security AnalysisAI

AI Category
APIs and Models
Risk Domain
N/A
OWASP Top 10 for LLMs 2025
None mapped
Classification Reason
Matched keywords: openai

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.
attack.mitre.org →
Derived from this CVE’s CWE(s) via the direct CWE→ATT&CK cross-walk.

CVEs Like This One

CVE-2026-25960Same product: Vllm Vllm
CVE-2026-34753Same product: Vllm Vllm
CVE-2026-24779Same product: Vllm Vllm
CVE-2026-27893Same product: Vllm Vllm
CVE-2024-11041Same product: Vllm Vllm
CVE-2025-24357Same product: Vllm Vllm
CVE-2025-47277Same product: Vllm Vllm
CVE-2025-30165Same product: Vllm Vllm
CVE-2025-32444Same product: Vllm Vllm
CVE-2025-29783Same product: Vllm Vllm

Affected Assets

vllm
vllm
0.3.0 — 0.22.0

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)
  • V4.1.3
  • V4.2.4
  • V1.5.3
  • V4.1.1

Mitigating Controls (NIST 800-53 r5) AI

AC-4 Information Flow Enforcement good match

Information flow enforcement directly stops trusted and untrusted data from being combined by applying rules that govern allowable data movements and combinations.

AC-16 Security and Privacy Attributes partial match

Associating explicit security attributes with data objects enables enforcement mechanisms that keep trust levels from being mixed inside structures.

SC-7 Boundary Protection partial match

Boundary protection at external interfaces can enforce consistent HTTP request/response parsing rules between intermediaries and endpoints.

SI-10 Information Input Validation partial match

Validating HTTP inputs at the intermediary prevents malformed messages from being interpreted inconsistently downstream.

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-01 mostly match
prevents

Configuration management can enforce uniform HTTP parsing rules across intermediaries, directly mitigating inconsistent interpretation.

PR.PS-06 mostly match
prevents

Secure-development practices (coding standards, reviews, validation) directly prevent mixing trusted and untrusted data inside the same structures.

DE.CM-01 partial match
prevents

Network monitoring can detect smuggling attempts via anomalous HTTP traffic or logs, while eliminating the inconsistency directly aids detection of such events.

ID.AM-03 partial match
prevents

Documented data-flow representations make trust boundaries explicit and help surface mixing of trusted/untrusted data.

PR.IR-01 partial match
prevents

Network protections can enforce consistent HTTP proxy/firewall behavior to block smuggling, and removing the weakness helps prevent unauthorized access via request smuggling.

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.

A.8.27 Secure system architecture and engineering principles mostly match
prevents

Secure architecture principles require explicit trust zones and data segregation, mitigating mixing of trusted/untrusted data.

A.8.28 Secure coding mostly match
prevents

Secure coding standards can enforce input validation and data tagging, but do not guarantee architectural separation.

A.8.29 Security testing in development and acceptance mostly match
finds

Security testing can detect HTTP request smuggling vulnerabilities in intermediary components.

A.8.20 Networks security partial match
degrades

Network security controls can enforce consistent HTTP parsing and proxy behavior that mitigates request smuggling.

A.8.21 Security of network services partial match
degrades

Secure network services include hardening proxies and gateways against inconsistent HTTP interpretation.

A.8.25 Secure development life cycle partial match
prevents

Secure SDLC practices require threat modeling and testing for HTTP parsing inconsistencies in intermediaries.

References