{"cve":"CVE-2026-115913","nvd":{"published":"2026-10-26T17:17:19.053","modified":"2026-10-26T18:34:17.050","status":"Awaiting Analysis","description":"vLLM is an inference and serving engine for large language models. Prior to 0.32.0, the Phi-4 multimodal audio encoder in phi4mm_audio.py and phi4mm_utils.py uses eager attention by default, allowing a remote client to submit one legal long audio clip that causes MultiHeadedAttention.forward_attention to materialize and retain full attention matrices in self.attn. The scaled dot-product attention branch does not materialize these matrices. The resulting quadratic GPU-memory use can cause an out-of-memory condition and terminate EngineCore for all users. This issue is fixed in version 0.32.0.","cvss_v4":null,"cvss_v3":{"version":"3.1","vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H","score":6.5},"cwes":["CWE-770"],"references":["https://github.com/vllm-project/vllm/commit/3d52578e13454de0bb8352cac988347cdcc89928","https://github.com/vllm-project/vllm/pull/60614","https://github.com/vllm-project/vllm/releases/tag/v0.32.0","https://github.com/vllm-project/vllm/security/advisories/GHSA-8397-8qf6-f2gm"],"products":[],"source":"nvd","synced_at":"2026-10-26T18:58:45.632391815+00:00"},"epss":null,"kev":null,"osv":[],"attribution":["This product uses the NVD API but is not endorsed or certified by the NVD.","EPSS scores: FIRST Exploit Prediction Scoring System, https://www.first.org/epss","Known exploited vulnerabilities: CISA Known Exploited Vulnerabilities Catalog, https://www.cisa.gov/known-exploited-vulnerabilities-catalog","Package advisories: OSV (https://osv.dev), from the GitHub Advisory Database, PyPA, Go, RustSec and other ecosystem databases, under CC-BY 4.0 or CC0","CVE records: the CVE Program's CVE List (https://www.cve.org), under its terms of use (https://www.cve.org/Legal/TermsOfUse). CVE is a registered trademark of The MITRE Corporation; this product is not affiliated with the CVE Program."]}