GPU Security: news, advisories and analysis

Latest coverage

• 

  Threats·Cloud & SaaSCritical

  vLLM Continuous Batching Engine Vulnerability Enables Cross-Tenant KV-Cache Extraction and Host Memory Corruption (CVE-2026-72110)

  vLLM issues critical advisory CVE-2026-72110 for its continuous batching inference engine, patching a PagedAttention memory calculation bug that leaked LLM KV-cache tokens.

  CST Editorial Desk11 Oct 20268 min read
• 

  Threats·Cloud & SaaSHigh

  NVIDIA Triton Inference Server Shared Memory Buffer Overflow Permits Local GPU Memory Corruption (CVE-2026-53112)

  NVIDIA issues a high-severity security bulletin for Triton Inference Server addressing a shared memory buffer overflow that permits arbitrary GPU memory corruption.

  Aisha Noor8 Oct 20268 min read
• 

  Threats·Cloud & SaaSCritical

  NVIDIA Triton Inference Server Memory Corruption Flaw Enables Remote Execution in GPU Clusters CVE-2026-34902

  NVIDIA Product Security advisory addresses a critical memory corruption defect in Triton Inference Server enabling code execution across enterprise AI clusters.

  Daniel Okafor6 Oct 20268 min read
• 

  Threats·Cloud & SaaSCritical

  vLLM CUDA Kernel Integer Overflow Leaks Cross-Tenant Inference Prompts

  An integer overflow flaw in vLLM PagedAttention CUDA kernels allows concurrent inference requests to read sensitive prompt tokens across multi-tenant GPU memory.

  Mei-Lin Chen27 Sep 20269 min read
• 

  Threats·Cloud & SaaSCritical

  vLLM Video Parser Buffer Overflow Enables Remote Root Code Execution

  vLLM maintainers and NIST disclose critical CVSS 9.8 flaw CVE-2026-22778 in multimodal video processing allowing unauthenticated remote root takeover of GPU clusters.

  Daniel Okafor20 Sep 202614 min read