The paper investigates how to optimally partition streaming inference models into megakernels and conventional graph components. It introduces a framework that predicts the best megakernel composition without full end‑to‑end runs, showing that multiple megakernels or a fully conventional graph often outperform a single megakernel across various models and batch sizes. The study defines four model categories based on megakernel stability and demonstrates significant latency and throughput gains on an H100 NVL GPU using a three‑megakernel design.
By Aditya Chatterjee, Anand Joseph X M, Sanjay Singh Chauhan
arXiv:2606. 28565v1 Announce Type: cross Abstract: As large language models (LLMs) move into production serving, practitioners must rapidly evaluate inference performance across diverse hardware, models, and serving parameters to meet cost and latency targets.
By Xiteng Yao, Taeho Kim, Hengzhi Pei, Xinle Liu, Kyle Ulrich, Leonard Lausen, Ashish Khetan, Xiang Song, George Karypis, Martin Herbordt
arXiv:2609.13585v1 Announce Type: cross
Abstract: Communication has become a bottleneck in distributed training and inference of large models. Overlapping communication with computation at the granul...
By Ziming Mao, Yihan Zhang, Shawn Wei Chew, Shuang Ma, Costin Raiciu, Yang Zhou, Scott Shenker, Ion Stoica
arXiv:2512. 22219v2 Announce Type: replace-cross Abstract: We introduce Mirage Persistent Kernel (MPK), the first compiler and runtime system that automatically transforms multi-GPU model inference into a single high-performance mega-kernel.
By Xinhao Cheng, Zhihao Zhang, Yu Zhou, Jianan Ji, Jinchen Jiang, Zepeng Zhao, Ziruo Xiao, Zihao Ye, Yingyi Huang, Ruihang Lai, Hongyi Jin, Bohan Hou, Mengdi Wu, Yixin Dong, Anthony Yip, Zihao Ye, Songting Wang, Wenqin Yang, Xupeng Miao, Tianqi Chen, Zhihao Jia
arXiv:2607. 23264v1 Announce Type: cross Abstract: Fine-grained, device-initiated communication lets persistent GPU kernels in distributed diffusion transformer (DiT) inference issue remote stores and overlap data movement with Tensor Core computation.
By Jianwen Xian, Zhiyuan Xu, Yuchen Li, Ziliang Lai, Kang He, Zhen Huang, Aichen Feng, Jinyan Chen, Yilin Zhang, Qinqin Chen, Chengru Song
arXiv:2607. 19539v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) architectures increase model capacity without proportionally increasing computation cost and have become a key building block for scaling large language models (LLMs) to trillion-parameter regimes.
By Minyu Cui, Anna Wingkvist, Morgan Ericsson
The paper investigates how a single SM utilization metric can misrepresent the true workload of large language model (LLM) inference on Nvidia Hopper GPUs. By profiling vLLM with FlashAttention‑3 and cuBLASLt on an H100 NVL across various phases (cold prefill, warm prefill, and decode) and varying sequence length and batch size, the authors replace the single utilization figure with eight detailed counter‑validated views. These views, tied to specific Nsight Compute counters or formulas, reveal how factors such as fragment fill, occupancy limits, stall signatures, wave quantization, and kernel selection create utilization gaps across four production models and six per‑layer kernel roles.
By Mohammad Siavashi, Gerald Q. Maguire Jr., Dejan Kostic, Marco Chiesa
arXiv:2607. 20475v1 Announce Type: new Abstract: Sampling in LLM inference comprises a combinatorial set of logit processing, token selection, and verification operations for speculative decoding.
By Pragaash Ponnusamy, Shivam Sahni, Jue Wang, Tri Dao
arXiv:2609.13592v1 Announce Type: cross
Abstract: GPU memory bandwidth and capacity limit throughput in large language model (LLM) inference. The GPU memory system consists of a primary tier of high-...
By Anish Saxena, Jae Hyung Ju, Hritvik Taneja, Po-An Tsai, Aamer Jaleel, Christos Kozyrakis, Moinuddin Qureshi
arXiv:2605. 25645v2 Announce Type: replace-cross Abstract: We present the first end-to-end demonstration of fine-tuning and serving Google's Gemma 4 31B model on TPU hardware, providing an empirical comparison of TPU and GPU platforms for large language model adaptation.
By Jatin Kishnani, Mayank Goel, Amit Singh, Pulkit Agrawal, Sairanjan Mishra
arXiv:2609.36070v1 Announce Type: cross
Abstract: AI accelerator systems are rapidly consolidating into scale-up architectures, where tens to thousands of GPUs communicate over high-bandwidth, single...
By Stuart H. Sul, Nash Brown, Henry Wildermuth, William Lin, Federico Cassano, Christopher R\'e
FluxMoE introduces an expert paging system that decouples Mixture-of-Experts (MoE) model experts from permanent GPU residency, allowing dynamic adaptation to available memory. By combining PagedTensor, a bandwidth‑balanced memory hierarchy, and a budget‑aware residency planner, FluxMoE streams expert weights on demand while keeping computations on GPUs. Experiments on GLM‑4.5 and Mixtral‑8×7B‑Instruct show significant throughput gains and reduced time‑per‑output‑token compared to existing inference engines, without compromising model quality.
By Qingxiu Liu, Yongchao He, Runhan Jiang, Zion Wang, Bohan Zhao, Mi Zhang, Patrick P. C. Lee