Login / Signup

Offline Energy-Optimal LLM Serving: Workload-Based Energy Models for LLM Inference on Heterogeneous Systems.

Grant WilkinsSrinivasan KeshavRichard Mortier
Published in: CoRR (2024)
Keyphrases
  • heterogeneous systems
  • energy consumption
  • response time
  • data management
  • database applications
  • distributed architecture