CVE-2026-105757

Published 5 October 2026 · tracked since 6 October 2026

Description

vLLM is an inference and serving engine for large language models. Prior to 0.30.0, structured-output request failures can escape request-scoped validation and reach the EngineCore fatal-error path. A per-request backend mismatch can re-raise a grammar compilation exception, padding produced by the

References

• NVD — National Vulnerability Database
• CVE.org record (MITRE)
• CISA Known Exploited Vulnerabilities catalog

Latest tracked vulnerabilities

• CVE-2026-105472 MEDIUM 6.3
• CVE-2026-104036 MEDIUM 5.8
• CVE-2026-105782 HIGH 7.5
• CVE-2026-105784 MEDIUM 4.6
• CVE-2026-105471 HIGH 7.3
• CVE-2026-105758 MEDIUM 5.3
• CVE-2026-105759 MEDIUM 5.9
• CVE-2026-105469 HIGH 7.3
• CVE-2026-105470 HIGH 7.3
• CVE-2026-105744 HIGH 7.5

→ Open the live CVE board · all tools