Dashboard / CVE-2026-115916

CVE-2026-115916

MEDIUM
MEDIUM Priority: 0.02 BB Relevance: 63/100
Published: 2026-10-26 Modified: 2026-10-28 First seen: 2026-10-29 Status: active

Attack Surface

Network Low Complexity No Interaction

CVSS Vector Breakdown

AV: Network
AC: Low
PR: Low
UI: None
S: Unchanged
C: None
I: None
A: High

Description

vLLM is an inference and serving engine for large language models. Prior to 0.32.0, vllm/tokenizers/hf.py computes max_token_id with an off-by-one upper bound, allowing a pre-tokenized prompt ID equal to the vocabulary size to pass validation. For a tensor-parallel size of one and an exact multiple-of-64 vocabulary, VocabParallelEmbedding.forward does not mask that value before F.embedding uses it, causing a CUDA device-side assertion and terminating the engine. This issue is fixed in version 0.32.0.

Weakness Classification (2)

Affected Software (1)

VendorProductVersion
vllm vllm < 0.32.0

References & Advisories (4)

Bug Bounty Relevance

63 / 100
✓ Network accessible
✓ Low complexity
✓ No user interaction

CVSS Score

6.5 / 10.0 v3.1
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H

EPSS — Exploitation Probability

0.31%
Raw Score 0.003120
Percentile 24.0%
Date 2026-10-29

EPSS Trend

2026-10-29
0.00312
2026-10-28
0.00305

Intelligence Sources

mitre ref →
nvd ref →

External Links