The paper presents a Pareto atlas of LLM inference optimizations, mapping cost, quality, and latency trade‑offs for Qwen2.5‑7B‑Instruct on L4, A100, and H100 GPUs. Using 54 measured configurations and a calibrated simulator, it identifies 18 of 36 setups on the Pareto frontier, showing that combined methods outperform single ones. Quality tests reveal that AWQ 4bit and FP8 weights offer significant latency reductions while largely preserving accuracy, but naive FP8 KV caching fails to answer any questions correctly.
By Srikanta Datta Tumkur, Jay Iyer, Mehar Simhadri, Sai Pavan Kumar, Sai Kapil Kumar, Ramesh Nampelly
arXiv:2609.14247v1 Announce Type: cross
Abstract: Bird-strike mitigation at wind farms requires identifying distant birds that span only tens of pixels, so the classifier's input resolution N is a de...
By Takeshi Nishikawa
TokenPowerSandbox is an evidence‑gated workflow that uses a CPU‑resident projector, brief GPU probes, full‑workload verification, and tamper‑evident provenance to predict energy usage of large language model serving. In experiments on an NVIDIA H100 80GB running Qwen2.5‑7B‑Instruct with vLLM, the method achieved energy MAPE of 6.23% and 7.35% on blind holdout and no‑refit confirmations, with high Spearman rank correlations. A predeclared TTFT gate demonstrated that energy accuracy alone cannot guarantee latency, as it passed at concurrency four but abstained below that level.
By Chenxu Niu
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. 02391v1 Announce Type: cross Abstract: Large Language Model (LLM) inference workloads are a rapidly growing contributor to data center energy consumption.
By Mauricio Fadel Argerich, Jonathan F\"urst, Marta Pati\~no-Mart\'inez
Large Language Model (LLM) inference workloads are a rapidly growing contributor to data center energy consumption. Optimizing these deployments requires matching specific LLMs to the most efficient GPUs, but operators currently lack the tools to do so without exhaustively profiling each combination.
arXiv:2603. 23640v2 Announce Type: replace-cross Abstract: Deploying large language models on-device for always-on personal agents demands sustained inference from hardware tightly constrained in power, thermal envelope, and memory.
By Pranay Tummalapalli, Sahil Arayakandy, Ritam Pal, Kautuk Kundan
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
The paper introduces a phase‑decoupled, model‑calibrated power controller for disaggregated large‑language‑model (LLM) serving, addressing the mismatch between GPU power settings and the distinct hardware regimes of prefill and decode stages. By calibrating separate power caps for each lane based on measured throughput‑latency cliffs, the authors achieve a 20.4% increase in tokens per joule with only a 3.5% rise in mean end‑to‑end latency on an 8‑node B200 cluster, outperforming NVIDIA’s Max‑Q profile. The approach also demonstrates consistent meeting of ITL‑p99 service‑level objectives across multiple MoE models and yields a 32.3% electricity savings over a sustained three‑day run.
By Jae Gon Kim, Donghoon Yoo, Hanyul Ryu, Sungho Ha, Juyeon Lee, Soojung Ryu
The paper evaluates NVIDIA’s Max‑Q inference profile on a disaggregated B200 GPU system for large language model (LLM) serving, finding modest gains (+8.6% tokens/J) but increased latency (+5.2%). It proposes a phase‑decoupled, model‑calibrated power controller that sets a latency‑guaranteed SM‑clock window for prefill and a calibrated power cap for decode, achieving a 20.4% tokens/J improvement with only a 3.5% latency increase on an 8‑node Qwen3‑Coder‑480B deployment. The approach outperforms vendor profiles on both energy and latency, and demonstrates significant long‑term electricity savings in MoE‑based serving.
arXiv:2607. 02518v1 Announce Type: cross Abstract: OpenClaw requests are dominated by long, tool-augmented prefixes, including system prompts, conversation history, and tool outputs fed back into the context window.
By Minjie Hua, Ning Wang, Peijun Yang, Kai Wang, Shiguo Lian
arXiv:2608. 15117v1 Announce Type: new Abstract: Analytical models of peak VRAM consumption for LLM inference decompose memory into weight-storage, KV-cache, and activation terms parameterized by step count, tool invocations, and context expansion.
By Anubhab Banerjee