PrivacyHigh
CVE-2026-73558: vLLM is an inference and serving engine for large language models. Prior to 0.27.0, an integer overf…
- Occurred
- Aug 13, 2026
- Published
- Aug 13, 2026
- Source
- NVD CVE
Summary
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.
Tags
- CVE
- CVE-2026-73558
- cvss:MEDIUM
Classification
Category confidence: 95%
Severity confidence: 90%
Rationale: Integer overflow vulnerability allows cross-user data leakage in vLLM inference engine, constituting unauthorized data access.
Classifier: Qwen/Qwen3.5-35B-A3B-FP8