Blink enables CPU-free LLM inference via SmartNIC offload and persistent GPU kernel, delivering up to 8.47x lower P99 TTFT, 3.4x lower P99 TPOT, 2.1x higher decode throughput, and 48.6% lower energy per token while remaining stable under CPU interference.
Title resolution pending
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 3verdicts
UNVERDICTED 3roles
background 1polarities
background 1representative citing papers
A vision for a cloud SmartNIC that hides Parquet decoding costs by offloading parsing and filters directly on the network datapath, backed by DuckDB performance estimates.
No single post-Moore technology replaces current HPC for plasma simulations, but FPGA-class accelerators offer near-term kernel offload, non-von Neumann architectures medium-term operator acceleration, and quantum computing long-term potential for warm dense matter microphysics.
citing papers explorer
-
Blink: CPU-Free LLM Inference by Delegating the Serving Stack to GPU and SmartNIC
Blink enables CPU-free LLM inference via SmartNIC offload and persistent GPU kernel, delivering up to 8.47x lower P99 TTFT, 3.4x lower P99 TPOT, 2.1x higher decode throughput, and 48.6% lower energy per token while remaining stable under CPU interference.
-
Should I Hide My Duck in the Lake?
A vision for a cloud SmartNIC that hides Parquet decoding costs by offloading parsing and filters directly on the network datapath, backed by DuckDB performance estimates.
-
Post-Moore Technologies for Plasma Simulation: A Community Roadmap
No single post-Moore technology replaces current HPC for plasma simulations, but FPGA-class accelerators offer near-term kernel offload, non-von Neumann architectures medium-term operator acceleration, and quantum computing long-term potential for warm dense matter microphysics.