CVE-2026-73558
Medium
Description
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.
SeverityMedium
CVSS5.3
Exploit probability—
ProductVllm
PublishedAug 13, 2026
First seenAug 13, 2026
Weakness typeCWE-190Integer overflow
How to fix
Update
- vllm vllm0.27.0 or later
- vllm0.27.0 or later
Affected products
| Product | Affected versions | Fixed in |
|---|---|---|
| vllm vllm | v < 0.27.0 | 0.27.0 |
| vllmPyPI | v < 0.27.0 | 0.27.0 |
Check your version
Choose the product and enter the version you run to see whether it falls in the affected range.
In the Intelligence 360 plan
Attack Surface Management
See the assets tied to your domains, the technologies they run and known vulnerabilities in one list.