CVE-2026-93592

7.5 HIGHpublic exploit available

Published 2026-09-18 · Updated 2026-09-22

AI analysis for this CVE has not been generated yet. Raw NVD data is shown below.

NVD description

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.

CVSS vector

CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H

Weaknesses

CWE-129

Public exploit & PoC references

https://github.com/vllm-project/vllm/security/advisories/GHSA-25q3-v2hm-8vpf
https://github.com/vllm-project/vllm/security/advisories/GHSA-25q3-v2hm-8vpf

All references

https://github.com/vllm-project/vllm/security/advisories/GHSA-25q3-v2hm-8vpf
https://www.vulncheck.com/advisories/vllm-before-0.28.0-denial-of-service-via-negative-token-id
https://github.com/vllm-project/vllm/security/advisories/GHSA-25q3-v2hm-8vpf

Source data: NVD (nvd.nist.gov), public domain. Exploit-DB.ai adds local AI analysis for defensive use only.

Related CVEs

MEDIUMCVE-2026-100654PoC
MEDIUMCVE-2026-100836PoC
LOWCVE-2026-102822PoC
UNSCOREDCVE-2026-16651PoC
MEDIUMCVE-2026-17413
HIGHCVE-2026-61695PoC
MEDIUMCVE-2026-62866PoC
MEDIUMCVE-2026-63420PoC

Related by shared AI tags and CWE weakness class. Browse the full archive.