hardware

arXiv:2606.13856v2 Announce Type: replace Abstract: Fine-tuning a vision-language-action model (VLA-JEPA) on a single GPU should be simple: load a pretrained checkpoint, run training, deploy. There is

Yet another vLLM fork thread here, but this time its for older INT8-centric hardware. This is a complete INT8 serving stack for Qwen3.8 27B based on vLLM, AITER, and a 27B GPTQ INT8 quant w/ DFlash2.

arXiv:2608.24154v1 Announce Type: cross Abstract: Real-world deployment of traffic surveillance systems is bottlenecked by geographic domain shift, in which models trained in one city underperform whe

arXiv:2602.23553v2 Announce Type: replace Abstract: Long-form video question answering (LVQA) requires answering natural-language queries about videos spanning minutes to hours, demanding temporal rea

arXiv:2608.24516v1 Announce Type: cross Abstract: Satellite-based distributed learning promises to train machine-learning models directly in orbit using massive, globally dispersed sensor data, thereb

arXiv:2608.24112v1 Announce Type: new Abstract: Modern text-to-image (T2I) models often have similar total scores but different strengths, making practical selection difficult. Fine-grained benchmarks

arXiv:2608.23807v1 Announce Type: new Abstract: Masked diffusion language models (dLLMs) can in principle generate text faster than autoregressive (AR) models, since they denoise many tokens at once.

arXiv:2608.23840v1 Announce Type: cross Abstract: Training large-scale AI models often outgrows a single data center, demanding sharded, multi-cluster, and decentralized training. However, the huge sp

The Information: Source: Nvidia has agreed to buy Hugging Face for 12.9B; the AI repository has had several potential suitors among its investors, including Salesforce — Nvidia has agreed to buy Huggi

Business Insider: Source: Nvidia has been in talks to acquire Hugging Face for more than $13B; Microsoft also met with Hugging Face, but talks are not ongoing — - The companies have not reached a deal

Brody Ford / Bloomberg: Sources: Anthropic has agreed to pay Nscale 45B over six years to rent about 460MW of power at a West Virginia data center using Nvidia's Vera Rubin chips — Anthropic PBC has a

The first phase of the generative artificial intelligence infrastructure boom was defined by a race for graphics processing units. The next phase will be defined by what happens after those GPUs arriv

SemiAnalysis: A detailed look at Jalapeño, OpenAI's ASIC developed with Broadcom in 16 months, which beat Nvidia, AMD, and Google chips on multiple top open-weight models — OpenAI's self-designed ASIC

arXiv:2608.23375v1 Announce Type: cross Abstract: Gumbel-based inference verification bounds LLM weight exfiltration by only forgiving token choices that plausibly arise from honest GPU nondeterminism

The artificial intelligence infrastructure market is crossing an important threshold. The conversation is shifting from acquiring graphics processing units to building complete AI factories that can g

arXiv:2608.21447v1 Announce Type: new Abstract: Structured-language models such as SceneScript reconstruct a scene as a short program of parametric commands, an inherently editable and semantically ex

AI factories are moving from ambitious plans toward production, but the path from graphics processing unit acquisition to usable systems remains a race against time. Neoclouds already have customers w

Cisco Systems Inc. today introduced an expansion to its current lineup of rack-scale artificial intelligence solutions for massive AI workloads in secure AI factories, addressing growing needs for mas

arXiv:2608.22115v1 Announce Type: new Abstract: Collaborative multimodal inference improves edge perception by combining observations from distributed sensing devices, but transmitting high-dimensiona

CUDA Python 1.0, released alongside CUDA 13.3, provides NVIDIA-maintained libraries and tools that expose the entire CUDA platform directly to Python through a single shared foundation. It adopts sema

For the past two years, the generative artificial intelligence boom has largely been defined by a single market. Hyperscalers have dominated the market with massive capital spending. According to Cred

arXiv:2608.21804v1 Announce Type: new Abstract: Graph-based point cloud registration achieves high robustness by identifying geometrically consistent correspondence sets, but constructing second-order

Had a lot of fun chatting again with my twin brother @dylan522p We went through lab economics over the next few years - the shift from inference to training as RSI draws near; and how Anthropic and Op

If you're looking for a weekend project, how about training your own text-to-speech model from scratch on your own GPU, and then running it on any device's CPU? We just open-sourced the entire Pocket

arXiv:2608.22506v1 Announce Type: new Abstract: Claim-level Uncertainty Quantification (UQ) aims to mitigate the lack of reliability of Large Language Models (LLMs) by evaluating the factuality of eac

NVIDIA is bringing the next wave of RTX gaming to the Gamescom conference running this week in Cologne, Germany, with support for new games, anti-cheat technologies and increased visual quality. Elect

arXiv:2608.21764v1 Announce Type: cross Abstract: Event-based vision has emerged as a promising paradigm for energy-aware artificial intelligence (AI), offering sparse, low-latency visual signals that

arXiv:2608.21806v1 Announce Type: cross Abstract: Computational resources are increasingly central to NLP research, but how closely reported GPU capability aligns with scholarly impact remains unclear

The Cisco Secure AI Factory with Nvidia has been extended into the rack-scale era, offering enterprises greater full-stack operational capabilities as a result. Designed to give customers a framework

Nvidia Corp. today announced the release of Jetson Orin Nano 2, a robotics computer “brain” for running artificial intelligence and frontier-level models at the edge. In the past months, foundational

arXiv:2608.22314v1 Announce Type: new Abstract: Many image processing methods such as corner detection, optical flow and iterative enhancement make use of image tensors. Generally, these tensors are e

OpenAI reports Jalapeño delivers 1.5–1.9× more AI work per watt and 1.7–3.6× lower latency vs comparison systems on InferenceX.

OpenAI Jalapeno is spicy Usually first generation chips aren't competitive, but OpenAI is beating Nvidia Blackwell and even Rubin This is huge news! We got to go into OpenAI's lab to dissect their new

OpenAI’s Jalapeño is a self‑designed ASIC for large‑language‑model inference that was taped‑out in about 16 months, developed with Broadcom and featuring HBM4 memory. Benchmarks on the InferenceX suit

arXiv:2608.23552v1 Announce Type: new Abstract: Language models are sequential processors, but long-horizon agency requires external information and computation beyond model weights and active context

Shadow Engine Recovery in NVIDIA Dynamo keeps a fully-initialized standby LLM engine on the same GPU, using the GPU Memory Service (GMS) to persist model weights and share them across engines without

Generalist AI Inc., a startup that develops artificial intelligence software for robots, has reportedly raised 200 million in funding. Axios today cited a source as saying that 8VC led the investment.

arXiv:2608.22526v1 Announce Type: new Abstract: We introduce RS^3-Prune, a training-free token-pruning recipe that instantiates as a small set of inference time hooks atop existing video object segmen

arXiv:2608.22300v1 Announce Type: new Abstract: Co-registration underlies nearly every multi-temporal and multi-sensor use of optical satellite imagery, and operational products still carry documented

Since announcing Jalapeño, our first custom inference chip, we’ve been testing it and the system around it. The results show a major advance: more intelligence from every watt and faster responses, de

arXiv:2509.04719v3 Announce Type: replace-cross Abstract: The widespread adoption of diffusion models for image generation necessitates efficient parallel inference to manage their substantial computa

arXiv:2608.21607v1 Announce Type: new Abstract: We investigate when a sparse nonnegative matrix can be recovered from a real-valued matrix of much lower rank by zeroing out its negative elements. The

arXiv:2608.22086v1 Announce Type: cross Abstract: We present SweepLSD, a line segment detector that reads the image exactly once and emits each segment within a few rows of its last pixel passing the

arXiv:2608.21555v1 Announce Type: cross Abstract: Modern machine learning compilers select tensor memory layouts to minimize execution cost under hardware constraints. Layout selection is global: an o

arXiv:2608.22007v1 Announce Type: cross Abstract: The components of a transformer communicate by writing to and reading from a shared residual stream, and mechanistic interpretability has mapped these

arXiv:2608.21887v1 Announce Type: cross Abstract: Runtime thermal management of high-performance chips depends on fast and accurate full-chip thermal maps. Conventional simulators typically estimate p