{"id":"CVE-2026-44222","aliases":["GHSA-hpv8-x276-m59f","PYSEC-2026-3409"],"url":"https://o3.security/vulnerability/CVE-2026-44222","summary":"vLLM: Remote DoS via Special-Token Placeholders","details":"vLLM is an inference and serving engine for large language models (LLMs). From 0.6.1 to before 0.20.0, there is a a Token Injection vulnerability in vLLM’s multimodal processing. Unauthenticated, text-only prompts that spell special tokens are interpreted as control. Image and video placeholder sequences supplied without matching data cause vLLM to index into empty grids during input-position computation, raising an unhandled IndexError and terminating the worker or degrading availability. Multimodal paths that rely on image_grid_thw/video_grid_thw are affected. This vulnerability is fixed in 0.20.0.","published":"2026-05-12T19:57:25.336Z","modified":"2026-08-07T11:31:35.028307085Z","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":"0.20.0"}],"fix":null,"references":[{"type":"ADVISORY","url":"https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/44xxx/CVE-2026-44222.json"},{"type":"ADVISORY","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-hpv8-x276-m59f"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-44222"},{"type":"REPORT","url":"https://github.com/vllm-project/vllm/issues/32656"}],"provenance":{"sources":["OSV.dev","FIRST.org (EPSS)"],"lastVerified":"2026-08-07T11:31:35.028307085Z"}}