CWE-444: Inconsistent Interpretation of HTTP Requests ('HTTP Request/Response Smuggling') in vllm-project vllm (CVE-2026-48746)
vLLM versions from 0.3.0 up to but not including 0.22.0 contain a critical vulnerability in the ASGI web servers and Starlette framework that allows authentication bypass of the OpenAI API AuthenticationMiddleware. This enables use of the API without providing the configured API key. The vulnerability is fixed in version 0.22.0.
AI Analysis
Technical Summary
The vLLM project versions >=0.3.0 and <0.22.0 are affected by a vulnerability classified as CWE-444 (Inconsistent Interpretation of HTTP Requests, also known as HTTP Request/Response Smuggling) that impacts the ASGI web servers and the Starlette framework. This flaw allows attackers to bypass authentication mechanisms in the OpenAI API AuthenticationMiddleware, permitting unauthorized API access without the required API key. The vulnerability has a CVSS v3.1 score of 9.1 (critical) with network attack vector, low attack complexity, no privileges required, no user interaction, unchanged scope, high confidentiality impact, no integrity impact, and high availability impact. The issue is resolved in vLLM version 0.22.0.
Potential Impact
Successful exploitation allows unauthenticated attackers to access the OpenAI API without providing the configured API key, resulting in a confidentiality breach and potential denial of service due to high availability impact. This compromises the security of the API and could lead to unauthorized usage or disruption of service.
Mitigation Recommendations
A fixed version 0.22.0 of vLLM is available and should be applied to remediate this vulnerability. Users should upgrade to vLLM version 0.22.0 or later. There is no indication that temporary mitigations or workarounds are provided by the vendor. Patch status is confirmed by the vendor advisory.
CWE-444: Inconsistent Interpretation of HTTP Requests ('HTTP Request/Response Smuggling') in vllm-project vllm (CVE-2026-48746)
Description
vLLM versions from 0.3.0 up to but not including 0.22.0 contain a critical vulnerability in the ASGI web servers and Starlette framework that allows authentication bypass of the OpenAI API AuthenticationMiddleware. This enables use of the API without providing the configured API key. The vulnerability is fixed in version 0.22.0.
CVSS v3.1
Score 9.1critical
Affected software
Run on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
The vLLM project versions >=0.3.0 and <0.22.0 are affected by a vulnerability classified as CWE-444 (Inconsistent Interpretation of HTTP Requests, also known as HTTP Request/Response Smuggling) that impacts the ASGI web servers and the Starlette framework. This flaw allows attackers to bypass authentication mechanisms in the OpenAI API AuthenticationMiddleware, permitting unauthorized API access without the required API key. The vulnerability has a CVSS v3.1 score of 9.1 (critical) with network attack vector, low attack complexity, no privileges required, no user interaction, unchanged scope, high confidentiality impact, no integrity impact, and high availability impact. The issue is resolved in vLLM version 0.22.0.
Potential Impact
Successful exploitation allows unauthenticated attackers to access the OpenAI API without providing the configured API key, resulting in a confidentiality breach and potential denial of service due to high availability impact. This compromises the security of the API and could lead to unauthorized usage or disruption of service.
Mitigation Recommendations
A fixed version 0.22.0 of vLLM is available and should be applied to remediate this vulnerability. Users should upgrade to vLLM version 0.22.0 or later. There is no indication that temporary mitigations or workarounds are provided by the vendor. Patch status is confirmed by the vendor advisory.
Technical Details
- Gcve Source
- db.gcve.eu
- Csaf Category
- csaf_base
- Csaf Version
- 2.0
- Publisher
- Bundesamt für Sicherheit in der Informationstechnik
- Advisory Id
- WID-SEC-W-2026-1974
- Cve Count
- 1
- Additional Cves
- []
- Cvss Version
- null
Threat ID: 6a32cff89f87a2db092d9c1a
Added to database: 06/17/2026, 16:48:56 UTC
Last enriched: 07/24/2026, 20:59:41 UTC
Last updated: 07/31/2026, 19:39:21 UTC
Views: 279
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.