#vllm
-
Unpatched LMCache Flaw Hands Remote Code Execution to Anyone on the Network
JFrog has disclosed a critical hole in LMCache, the open-source accelerator for LLM servers: a single network message can run code on the cache server — as root on the project's official images. No fix exists yet.
-
Can an AI Take Over the Server Through Inference-Engine Vulnerabilities?
An essay by Boyd Kane on how a malicious language model could gain control of the very machine hosting its own weights by exploiting bugs in engines like vLLM and SGLang.
-
Why a Local LLM Seems Dumber Than It Really Is
A poster on the Level1Techs forum ran a series of experiments showing that the very same model weights produce different tokens depending on the hardware and software — and that there's some very concrete math behind it.
-
Baidu Releases vLLM-Kunlun — a Plugin for Running LLMs on Kunlun XPU Accelerators
Baidu has open-sourced vLLM-Kunlun, a pluggable module that lets you run the popular vLLM inference engine on its own Kunlun XPU accelerators without patching vLLM's source code.