Dissecting GPU Utilization for LLM Inference on Nvidia Hopper

arXiv:2609.12923 · cs.PF, cs.AR, cs.DC, cs.LG · Submitted 2026-09-11 · Read on arXiv

cs.PF, cs.AR, cs.DC, cs.LG

Submitted: 2026-09-11

Updated: 2026-09-11

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Decode — 3× in this paper · explained in 2 episodes
• Quantization — 1× in this paper · explained in 15 episodes

Related papers

• HybridQC: Hardware-Grounded Simulation of Tightly Integrated Hybrid Quantum-Classical Systems
• Reproducible LLM Inference Benchmarking: A Sequential Isolation Protocol for Regression Testing
• A Controlled Study of Memory Hierarchy Transitions in Quantum Circuit Simulation on Apple M4 Pro Unified Memory Architecture
• The Caf'e in Amsterdam: When the Incumbent Becomes the Oracle