CVE-2026-53923

vLLM is an inference and serving engine for large language models (LLMs). From 0.5.5 until 0.23.1rc0, integer truncation of tensor dimensions in vLLM's GGUF dequantize kernels (csrc/quantization/gguf/gguf_kernel.cu) causes partial tensor processing. The output tensor is allocated at full size via torch::empty (uninitialized memory), but the dequantize CUDA kernel processes only a truncated number of elements. The unfilled portion of the output tensor retains whatever was previously in GPU memory. In multi-tenant inference deployments, this residual GPU memory may contain tensor data from other users' inference requests, constituting information disclosure. This vulnerability is fixed in 0.23.1rc0.

Affected products

Vllm

Vllm

< 0.23.1

Fix

Available

CVSS 3.1

7.5 HIGH

EPSS

0.3% (20th percentile)

Weakness

CWE-681, CWE-200

NVD status

Analyzed

Published

2026-06-22

No exploit for CVE-2026-53923 has been indexed yet. Our index is built from live traffic and upstream syncs, so this page can only say what it knows — not that no exploit exists.