vLLM Affected by Denial of Service via Unbounded Frame Count in video/jpeg Base64 Processing (vllm-project vllm)
At a glance
- Source
- NVD
- EPSS
- EPSS 0.46 % · percentile 40 ()
- State
- PUBLISHED
- Published
- Updated
- Assigner
- GitHub_M
- NVD status
- Modified
Description
vLLM is an inference and serving engine for large language models (LLMs). From 0.7.0 to before 0.19.0, the VideoMediaIO.load_base64() method at vllm/multimodal/media/video.py splits video/jpeg data URLs by comma to extract individual JPEG frames, but does not enforce a frame count limit. The num_frames parameter (default: 32), which is enforced by the load_bytes() code path, is completely bypassed in the video/jpeg base64 path. An attacker can send a single API request containing thousands of comma-separated base64-encoded JPEG frames, causing the server to decode all frames into memory and crash with OOM. This vulnerability is fixed in 0.19.0.
Weaknesses (CWE)
- CWE-770 — Allocation of Resources Without Limits or Throttling
References
- https://github.com/vllm-project/vllm/security/advisories/GHSA-pq5c-rjhq-qp7p
- https://access.redhat.com/security/cve/CVE-2026-34755
- https://bugzilla.redhat.com/show_bug.cgi?id=2455403
- https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-34755.json
- https://access.redhat.com/errata/RHSA-2026:36005
- https://access.redhat.com/errata/RHSA-2026:36006
- https://access.redhat.com/errata/RHSA-2026:57380
- https://access.redhat.com/errata/RHSA-2026:57389
- https://access.redhat.com/errata/RHSA-2026:57390
- https://access.redhat.com/errata/RHSA-2026:57387
- https://access.redhat.com/errata/RHSA-2026:59151
- https://access.redhat.com/errata/RHSA-2026:59144
Scores
| Version | Score | Vector | Scored by |
|---|---|---|---|
| CVSS 3.1 | MEDIUM 6.5 | CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H | GitHub_M |
| CVSS 3.1 | MEDIUM 6.5 | CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H | redhat-SADP |
| CVSS 3.1 | MEDIUM 6.5 | CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H | NVD · security-advisories@github.com (Secondary) |
| CVSS 3.1 | MEDIUM 6.5 | CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H | NVD · 0b0ca135-0b70-47e7-9f44-1890c2a1c46c (Secondary) |
Advisories
- GHSA-pq5c-rjhq-qp7p — vLLM: Denial of Service via Unbounded Frame Count in video/jpeg Base64 Processing
- PYSEC-2026-144 — vLLM is an inference and serving engine for large language models (LLMs). From 0.7.0 to before 0.19.0, the VideoMediaIO.load_base64() method at vllm/multimodal/
European references
- ENISA EUVD · EUVD-2026-19350
- GCVE · GCVE-0-2026-34755
Affected configurations (NVD CPE)
cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*≥ 0.7.0 < 0.19.0
Affected products
| Product | Versions |
|---|---|
| vllm-project · vllm | affected >= 0.7.0, < 0.19.0 |
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 |