Monday 16 November
Global

Your region, kept on this device.

Vulnerability · Vllm Vllm

CVE-2026-115913

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 score (v3.1)6.5Medium
Fixed in 0.32.0
Not on CISA KEV list

Am I affected?

Sign in and add the software you use to your watchlist: we check it against every new vulnerability.

Show me how to fix it

Timeline

  1. Published by NVD
  2. Last updated by NVD

Affected products

ProductVersions
Vllm Vllmfrom 0.31.0 to before 0.32.0

Sources: NVD and CISA known exploited vulnerabilities catalogue.