Skip to content
CORVASEC — Knowledge Defends
Vulnerability Explorer›CVE-2026-105754
VULNERABILITY

CVE-2026-105754

Common Vulnerabilities and Exposures record

MEDIUM

Summary

vLLM is an inference and serving engine for large language models. Prior to 0.30.0, the /inference/v1/generate endpoint in the disaggregated scale-out path accepts caller-supplied tensors in the features.kwargs_data field, cache identifiers in the features.mm_hashes field, ranges in the features.mm_placeholders field, and wire-selected multimodal field processors without rebinding them to the active model renderer contract. Forged grid geometry, field types, or non-positive placeholder lengths can terminate the shared EngineCore; when an attacker knows or can induce a victim's content hash, forged cache hashes can poison or retrieve cross-request encoder-cache state; and dropped sparse placeholder masks can alter replayed transport semantics. This issue is fixed in version 0.30.0.

Risk

CVSS
6.5
Severity
MEDIUM
Vector
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H

Timeline

Published
Oct 5, 2026 23:17
Modified
Oct 8, 2026 03:16
Last observed
Oct 13, 2026 17:01

CWE

CWE-20CWE-617CWE-639CWE-668CWE-704CWE-1284

Affected product identifiers

cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*

References