CVE-2026-73558

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.

References

• https://github.com/vllm-project/vllm/commit/451227cb3ff07989698fed982c2d3e4300257924
• https://github.com/vllm-project/vllm/issues/42860
• https://github.com/vllm-project/vllm/pull/49660
• https://github.com/vllm-project/vllm/releases/tag/v0.27.0
• https://github.com/vllm-project/vllm/security/advisories/GHSA-7m6h-x95x-82q5