Unauthenticated API access via malformed Host header
Published Jun 22, 2026 · Updated Jun 22, 2026
Authentication bypass in vLLM allows remote attackers to access protected OpenAI API endpoints through a malformed Host header. AuthenticationMiddleware checks a URL path reconstructed by Starlette from the unvalidated Host header, while FastAPI routes on the actual request path. The flaw affects 0.3.0 and later releases before 0.22.0 when API-key authentication is enabled, the API is attacker-reachable, and no RFC-conforming reverse proxy rejects the header.
Summary
What happened
Authentication bypass in vLLM allows remote attackers to access protected OpenAI API endpoints through a malformed Host header. AuthenticationMiddleware checks a URL path reconstructed by Starlette from the unvalidated Host header, while FastAPI routes on the actual request path. The flaw affects 0.3.0 and later releases before 0.22.0 when API-key authentication is enabled, the API is attacker-reachable, and no RFC-conforming reverse proxy rejects the header.
The record
- CVE
- CVE-2026-48746
- Published
- Jun 22, 2026
- Updated
- Jun 22, 2026
- Vendor
- vLLM Project
- Product
- vLLM
- Classifications
- CWE-444, CWE-501, T1190
- Attack vector
- network
- Privileges
- unauthenticated
Timeline
How it unfolded
- Jun 22, 2026CVE publishedPublication date reported by the CVE source.
- Jun 22, 2026Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=>= 0.3.0, < 0.22.0
What conditions does exploitation require?
What is affected?
Published CVSS scores
CVSS describes severity. EPSS estimates exploitation probability.
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
- Malformed Host-header authentication-bypass proof of conceptproof of concept · demonstrated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo