NetSight
hoch7.5

CVE-2026-53923

vLLM: GGUF dequantize kernel int truncation exposes uninitialized GPU memory in multi-tenant serving

Auswirkung

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.

Ein Angreifer kann die Lücke über das Netzwerk ausnutzen, ohne Anmeldung, ohne Zutun eines Nutzers. Bei Erfolg: hohes Risiko, dass Daten ausgelesen werden.

Betroffene Pakete

ÖkosystemPaketBetroffenBehoben in
pipvllm>= 0.5.5, < 0.24.00.24.0

Betroffene Produkte · Vllm

ProduktBetroffene Versionen
Vllm
ab 0.5.5 vor 0.23.1

Schwachstellenklasse

Referenzen