Skip to content
vLLMPYSEC-2026-3997

vLLM: unauthenticated attacker could crash the engine

UnratedCVE-2026-93592 · Published Sep 18, 2026 · updated Sep 29, 2026

vLLM versions before 0.28.0 fail to validate the lower bound of token IDs in the /v1/embeddings and /pooling endpoints, allowing unauthenticated attackers to crash the engine by submitting negative token IDs. A single request with a negative token ID triggers a CUDA device-side assertion that poisons the GPU context, causing all subsequent requests to fail until the process restarts.

Source advisory

Affected versions

PackageAffectedFixed in
vllm
PyPI
< 0.28.00.28.0
Details and references
CVSS 4.0
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X
Severity from
no source yet
Also known as
CVE-2026-93592, GHSA-25q3-v2hm-8vpf

More vLLM advisories

All vLLM

Critical advisories by email

Wednesdays: the week’s critical and high advisories in the AI and data stack, with the fixed versions. Only in weeks that have some.

Double opt-in. Unsubscribe any time.