Architecture-Aware LLM Inference Optimization on AMD Instinct GPUs: A Comprehensive Benchmark and Deployment Study

Documenti analoghi

• Rotary GPU: Exploring Local Execution Paths for Large Mixture-of-Experts Models Under Limited GPU Memory
  di: Jo, Myeong Jun
  Pubblicazione: (2026)
• Predictive Multi-Tier Memory Management for KV Cache in Large-Scale GPU Inference
  di: Ganjihal, Sanjeev Rao
  Pubblicazione: (2026)
• Nanoscaling Floating-Point (NxFP): NanoMantissa, Adaptive Microexponents, and Code Recycling for Direct-Cast Compression of Large Language Models
  di: Lo, Yun-Chen, et al.
  Pubblicazione: (2024)
• AIvailable: A Software-Defined Architecture for LLM-as-a-Service on Heterogeneous and Legacy GPUs
  di: Antunes, Pedro, et al.
  Pubblicazione: (2025)
• Evaluating Emerging AI/ML Accelerators: IPU, RDU, and NVIDIA/AMD GPUs
  di: Peng, Hongwu, et al.
  Pubblicazione: (2023)