arXiv:2607. 06202v1 Announce Type: cross Abstract: The deployment of Mixture-of-Experts (MoE) models on production high-bandwidth superpods, such as NVIDIA's NVL72/576 and Huawei's CloudMatrix384, introduces critical challenges beyond raw interconnect bandwidth.
arXiv:2606. 19025v1 Announce Type: cross Abstract: Pre-training Large Language Models (LLMs) typically demands large-scale infrastructure with tightly coupled hardware accelerators.
arXiv:2606. 10493v1 Announce Type: cross Abstract: Local deployment of large Mixture-of-Experts (MoE) models falls short of the service quality achieved in cloud-scale environments, even under low-concurrency workloads.
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.
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...
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.
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...
arXiv:2504. 08791v3 Announce Type: replace-cross Abstract: On-device inference offers privacy, offline use, and instant response, but consumer hardware restricts large language models (LLMs) to low throughput and capability.
The paper presents a method for distributing large language model inference across multiple Intel AI PCs by splitting the model into pipeline shards, each pre‑compiled into an OpenVINO graph. Three key techniques—beam_idx Gather to enable GPU optimizations, speculative decoding on stateful models, and interleaved micro‑batching—allow a two‑node Llama 3.1 8B INT4 pipeline to serve two users at 1.79× the throughput of a single‑node model, while a four‑node deployment can run a 70B model that no single PC can hold. The authors provide code, benchmark logs, and reproduction scripts on GitHub.
The paper introduces Federation of Experts (FoE), a new architecture that reorganizes the mixture-of-experts (MoE) block in transformer layers into multiple MoE clusters. Each cluster handles a single KV head, and expert parallelism is applied within clusters while a sum operation synchronizes post‑attention residuals across clusters. FoE eliminates all‑to‑all communication on a single GPU and limits it to intra‑node communication in multi‑node setups, leading to significant reductions in inference latency and throughput improvements on LongBench.
arXiv:2607. 24787v1 Announce Type: new Abstract: Sparse Mixture-of-Experts (MoE) models expand foundation model capacity through conditional expert activation, but their full expert pools remain difficult to deploy under limited accelerator memory.
ThunderEP is a new communication design for expert parallelism on PCIe-connected consumer GPUs that eliminates relay hops, uses DMA engines to avoid GPU compute contention, and reduces CPU polling overhead. Integrated into vLLM, it outperforms NCCL on RTX 4090 and RTX 5090 GPUs, delivering average speedups of 2.00× for dispatch, 1.53× for combine, and up to 1.66× end‑to‑end over existing MoE inference frameworks.
arXiv:2506.10911v2 Announce Type: replace Abstract: Training large language models is generally done on clusters containing thousands of accelerators, communicating over a high-bandwidth interconnect...