HorizonServe coordinates which requests enter the shared stage of an omni model and how many GPU processors that stage gets, lifting SLO attainment by up to 4.9x to 7x in tests.
2026.NVIDIA RTX PRO 6000 Blackwell Workstation Edition.https://www.nvidia.com/en-us/products/workstations/ professional-desktop-gpus/rtx-pro-6000/
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.DC 1years
2026 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
HorizonServe: Coordinating Request Scheduling with GPU Sharing for Omni-Model Serving
HorizonServe coordinates which requests enter the shared stage of an omni model and how many GPU processors that stage gets, lifting SLO attainment by up to 4.9x to 7x in tests.