8.2 High

CVE-2026-116172 vLLM 0.22.0 before 0.24.0 DoS via Rust gRPC CandidateTokens.top_n

Description

vLLM from 0.22.0 before 0.24.0 contains an improper input validation vulnerability in the Rust gRPC converter that allows remote attackers to bypass max_logprobs limits via CandidateTokens.top_n. Attackers can send Generate or GenerateStream requests with oversized or integer-wrapping top_n values to trigger torch.topk failures that terminate EngineCore.

CVSS 4.0 detail

CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N

Exploitable remotely over the network, without an account and without any user action; an attacker can seriously compromise confidentiality, integrity or availability.

Attack vector: Network
Attack complexity: High
Attack requirements: Present
Privileges required: None
Interaction: None
Confidentiality: None
Integrity: None
Availability: High
Confidentiality of other systems: None
Integrity of other systems: None
Availability of other systems: None

Affected products and versions

Product	Affected versions	Fixed in
vllm 0.24	earliest → < 0.24.0	0.24.0
vllm 0.22	0.22.0 → < 0.24.0	0.24.0

References

GitHub Security Advisory (GHSA-rwfw-xvg3-3937) github.com
VulnCheck Advisory: vLLM 0.22.0 before 0.24.0 DoS via Rust gRPC CandidateTokens.top_n www.vulncheck.com

Timeline

Published October 25, 2026
Last modified November 4, 2026

The 121 CVEs of vllm →

CVE-2026-115918 vLLM is an inference and serving engine for large language models 6.5 Medium CVE-2026-115917 vLLM is an inference and serving engine for large language models 3.4 Low CVE-2026-115916 vLLM is an inference and serving engine for large language models 6.5 Medium CVE-2026-115915 vLLM is an inference and serving engine for large language models 8.1 High

Track this CVE