vLLM: GLMGA video sampling permits request-driven CPU and memory exhaustion (vllm-project vllm)
At a glance
- Source
- NVD
- EPSS
- EPSS 0.30 % · percentile 22 ()
- State
- PUBLISHED
- Published
- Updated
- Assigner
- GitHub_M
- NVD status
- Analyzed
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.
Weaknesses (CWE)
- CWE-400 — Uncontrolled Resource Consumption
References
Scores
| Version | Score | Vector | Scored by |
|---|---|---|---|
| CVSS 3.1 | MEDIUM 5.3 | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L | GitHub_M |
| CVSS 3.1 | MEDIUM 5.3 | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L | NVD · security-advisories@github.com (Secondary) |
Advisories
- GHSA-58v5-2m8f-94pr — vLLM: GLMGA video sampling permits request-driven CPU and memory exhaustion
- PYSEC-2026-4307 — vLLM: GLMGA video sampling permits request-driven CPU and memory exhaustion
European references
- ENISA EUVD · EUVD-2026-92851
- GCVE · GCVE-0-2026-105760
Affected configurations (NVD CPE)
cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*≥ 0.23.0 < 0.30.0
Affected products
| Product | Versions |
|---|---|
| vllm-project · vllm | affected >= 0.23.0rc2, < 0.30.0 |
Sources
| Source | Licence |
|---|---|
| CVE List (CVEProject cvelistV5) | CVE Program Terms of Use |
| EPSS (FIRST.org) | FIRST — attribution requested |
| ENISA EUVD | ENISA — reproduction authorised, source acknowledged |
| NVD (NIST) via fkie-cad/nvd-json-data-feeds | Public domain — this product uses data from the NVD API but is not endorsed or certified by the NVD |