Preprint
Aug 2026
Hydra: Phase-Aware Workload Characterization of LLM Inference across Edge SoC Generations, Backends, and Quantization Levels
Hydra is presented, a common-schema, phase-aware workload characterization framework for LLM inference on edge SoCs that enables reproducible, phase-aware characterization of edge LLM inference.
Amir Taherin, Sana Taghipour Anvari, Charles Amante et al.
· 1 citation