CVE-2026-105754
- Source
- https://cve.org/CVERecord?id=CVE-2026-105754
- Import Source
- https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2026-105754.json
- JSON Data
- https://api.osv.dev/v1/vulns/CVE-2026-105754
- Aliases
- Downstream
CGA (8)
- Published
- 2026-10-05T22:46:03Z
- Modified
- 2026-10-07T02:47:27Z
- Severity
-
- 6.5 (Medium) CVSS_V3 - CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H CVSS Calculator
- Summary
- vLLM: Scale-out disaggregated multimodal transport trusts caller-supplied features
- Details
-
vLLM is an inference and serving engine for large language models. Prior to 0.30.0, the /inference/v1/generate endpoint in the disaggregated scale-out path accepts caller-supplied tensors in the features.kwargs_data field, cache identifiers in the features.mm_hashes field, ranges in the features.mm_placeholders field, and wire-selected multimodal field processors without rebinding them to the active model renderer contract. Forged grid geometry, field types, or non-positive placeholder lengths can terminate the shared EngineCore; when an attacker knows or can induce a victim's content hash, forged cache hashes can poison or retrieve cross-request encoder-cache state; and dropped sparse placeholder masks can alter replayed transport semantics. This issue is fixed in version 0.30.0.
- Database specific
{ "cna_assigner": "GitHub_M", "cwe_ids": [ "CWE-1284", "CWE-20", "CWE-617", "CWE-639", "CWE-668", "CWE-704" ], "osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/105xxx/CVE-2026-105754.json" }- References
-
- https://github.com/vllm-project/vllm/releases/tag/v0.30.0
- https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/105xxx/CVE-2026-105754.json
- https://github.com/vllm-project/vllm/security/advisories/GHSA-ph72-cqr5-qpp7
- https://nvd.nist.gov/vuln/detail/CVE-2026-105754
- https://github.com/vllm-project/vllm/commit/1970f3ed4be7fa8620e4ddc4a12c36a8384cfc27
- https://github.com/vllm-project/vllm/pull/51898