{"id":"CVE-2026-34755","aliases":["GHSA-pq5c-rjhq-qp7p","PYSEC-2026-144"],"url":"https://o3.security/vulnerability/CVE-2026-34755","summary":"vLLM Affected by Denial of Service via Unbounded Frame Count in video/jpeg Base64 Processing","details":"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.","published":"2026-04-06T15:38:53.201Z","modified":"2026-08-07T11:31:12.594309318Z","cvss":{"score":6.5,"severity":"MEDIUM","vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H"},"epss":null,"cisaKev":null,"exploitsKnown":0,"affectedPackages":[{"ecosystem":"PyPI","name":"vllm","fixedVersion":"0.19.0"}],"fix":null,"references":[{"type":"WEB","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-34755.json"},{"type":"ADVISORY","url":"https://access.redhat.com/errata/RHSA-2026:36005"},{"type":"ADVISORY","url":"https://access.redhat.com/errata/RHSA-2026:36006"},{"type":"ADVISORY","url":"https://access.redhat.com/security/cve/CVE-2026-34755"},{"type":"ADVISORY","url":"https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/34xxx/CVE-2026-34755.json"},{"type":"ADVISORY","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-pq5c-rjhq-qp7p"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-34755"},{"type":"REPORT","url":"https://bugzilla.redhat.com/show_bug.cgi?id=2455403"}],"provenance":{"sources":["OSV.dev","FIRST.org (EPSS)"],"lastVerified":"2026-08-07T11:31:12.594309318Z"}}