CVE-2026-105760

vLLM: GLMGA video sampling permits request-driven CPU and memory exhaustion

  • Medium 5.3

Description

vLLM is an inference and serving engine for large language models. Prior to 0.30.0, a caller can use the request-level media_io_kwargs field to select the GLMGA video backend and supply large values for the fps and max_frames options without a strict work ceiling. GLMGA constructs and deduplicates an attacker-sized pre-decode frame-index list, allowing a compact request and tiny valid video to consume disproportionate CPU time and memory in the shared media-loading executor. This issue is fixed in version 0.30.0.

Key facts

Published
5 Oct 2026
Last changed
6 Oct 2026
Assigned by
GitHub_M
EPSS
0.297% likely to be exploited in the next 30 days, higher than 22.625% of CVEs (scores for 17 Nov 2026)
SSVC (CISA)
Exploitation: none; Automatable: yes; Technical impact: partial

Severity scores

Each score is shown as its source gave it. They are not averaged or merged.

  • 5.3CVSS 3.1, Mediumfrom the CNA (GitHub_M)CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L

Historical context and remediation

This reference retains the original CVE publication date. A recent record update does not mean a new attack occurred.

No separately sourced impact summary is available on this page yet. Consult the description, version data and dated exploitation evidence.

No explicit remediation statement is available on this page yet. Consult the original record and vendor references before choosing a fix.

Version statuses below come from the CNA. An “unaffected” version is not necessarily a currently supported release or proof of a patch date. EPSS is a dated risk estimate, not a count of historical attacks.

Read the original CVE record.

Affected products

VendorProductVersions
vllm-projectvllm
  • >= 0.23.0rc2, < 0.30.0 (affected)

Timeline

  1. 5 Oct 2026 Published

Weakness

References