Skip to content

Main Navigation

UC Berkeley Sky Computing Lab
  • People
  • Projects
  • Publications
  • News
  • Events
  • Sponsors
  • Contact
  • DARE

Announcing vLLM-Omni: Easy, Fast, and Cheap Omni-Modality Model Serving

Posted on December 12, 2025 by Ivan Ortega

Post navigation

 Tracing Hanging and Complicated GPU Kernels Down To The Source CodeStreamlined multi-node serving with Ray symmetric-run 
  • People
  • Projects
  • Publications
  • News
  • Events
  • Sponsors
  • Contact
  • DARE
...




Copyright 2022 - SKY Computing UC Berkeley Electrical Engineering and Computer Science Department (EECS)