Zurück zur CVE-Übersicht
CVE-2026-105760
MEDIUM(5.3)CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L
Risk Signal Score13/100 — NIEDRIG
- CVSS 5.3 — Mittel
Erwähnungen (letzte 60 Tage)
Artikel
EPSS-Score
0.3%
Exploit-Wahrscheinlichkeit (30 Tage)
CVSS Score
5.3
Technische Schwere
Beschreibung
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.
Betroffene Produkte
- cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*
GitHub Advisories
GHSA-58v5-2m8f-94prMEDIUM
vLLM: GLMGA video sampling permits request-driven CPU and memory exhaustion
pip/vllm→ 0.30.0
GitHub Advisory