Medium [CVE-2026-73558] Cross-User Data Leakage via Integer Overflow

vLLM is an inference and serving engine for large language models. Prior to 0.27.0, an integer overflow in blockIdx.x * 2 * d in activation_kernels.cu can cause act_and_mul_kernel to consume another batched user's input, allowing a request processed in the same inference batch to receive a partial or complete copy of another user's inference result.

This issue is fixed in version 0.27.0. An integer overflow vulnerability in the `act_and_mul_kernel` kernel allows a remote attacker to cause cross-user data leakage.

This could lead to the unauthorized disclosure of sensitive information. This is particularly relevant in multi-tenant environments where vLLM processes requests concurrently from different users.

Red Hat severity: Moderate — CVSS 5.3 (CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:U/C:H/I:N/A:N). Weakness: CWE-190.

Affected Red Hat products: Red Hat AI Inference Server; Red Hat Enterprise Linux AI (RHEL AI) 3; Red Hat OpenShift AI (RHOAI). Red Hat does not currently list a fixing RHSA for this CVE.