hardware

Together AI serves the two fastest STT models measured by @ArtificialAnlys NVIDIA Parakeet-TDT 0.6B v3 can transcribe 20 hours of speech in under 10 seconds. This deep dive shows the systems work behi

arXiv:2605.30351v1 Announce Type: cross Abstract: Long-rollout causal video diffusion has converged on a fixed-size sliding-window KV cache, with recent progress innovating within this layout by chang

This Reddit discussion in the StableDiffusion community addresses recommendations for affordable, used or refurbished laptops equipped with GPUs suitable for running open-source image generation model

PCMag: What to expect at Computex 2026: AI chips, budget PCs competing with the MacBook Neo, Nvidia entering laptop SoC market with the rumored N1X chip, and more — The theme for Computex 2026, accord

arXiv:2605.27441v1 Announce Type: cross Abstract: Query understanding in large-scale industrial search systems is typically implemented as a cascade of disparate, task-specific components. While indiv

and see, perhaps further reflection of the death of tokenmaxxing The price to rent an Nvidia H200 just collapsed from 7/hr to 4/hr in three weeks. A -40% drop in the cost of the single most strategic

Gary Marcus discusses declining prices for NVIDIA's H100 and H200 AI chips, suggesting that inflated valuations and returns on investment for AI hardware customers are becoming unsustainable as the ma

arXiv:2605.27678v1 Announce Type: new Abstract: Foundation model training is becoming multimodal, from post-training pipelines to large-scale pretraining. As modality coverage broadens, context window

Intel introduced the Arc G-Series processors, a new family designed for next-generation handheld gaming systems, launching with Arc G3 and Arc G3 Extreme processors running Windows 11 based on Intel C

This article likely explores practical applications of image processing for IoT devices deployed across hybrid cloud architectures, discussing how organizations can leverage both on-premises and cloud

arXiv:2605.28213v1 Announce Type: new Abstract: LLM-based agents are increasingly used to generate GPU kernels, but they often know what optimizations to try without knowing when those optimizations a

arXiv:2605.27787v1 Announce Type: cross Abstract: Multi-agent systems (MAS) have substantially advanced autonomous software engineering (SWE), but their growing inference energy demands raise sustaina

Robotics is entering a new phase: moving from controlled demos and scripted automation toward generalizable, reliable embodied autonomy in the real world. At the International Conference on Robotics a

NVIDIA has released GLM-5.1-NVFP4, a quantized version of the GLM-5.1 model, now available on Hugging Face. The model appears to use NVFP4 (NVIDIA's floating-point 4-bit) quantization format, designed

arXiv:2605.28717v1 Announce Type: new Abstract: Modern datacenter RDMA is bottlenecked at the network interface, not the wire. A NIC running RoCE or InfiniBand holds per-connection state for every (ap

Reuters: Source: the Shanghai Futures Exchange is in the early stages of designing futures contracts for AI tokens; US exchanges are set to launch GPU compute futures — China is designing a futures ma

arXiv:2605.28764v1 Announce Type: new Abstract: Vast quantities of compute (GPU cycles on personal workstations, idle inference servers, and edge devices between jobs) go unused because no incentive-a

The HF science team just made async RL weight sync ~100x cheaper on bandwidth, and you don't need a shared cluster anymore. The problem: every RL step, the trainer typically has to sync fresh weights

The price to rent an Nvidia H200 just collapsed from 7/hr to 4/hr in three weeks. A -40% drop in the cost of the single most strategic asset in tech. When the underlying commodity that powers your ent

today is (potentially) a great day for the GPU poors if DiffusionBlocks works on fine-tuning existing models, then literally any reasonable consumer GPU can do LLM fine-tuning will make a video on thi

Today, we're releasing LFM2.5-8B-A1B, a device-optimized model designed to power real-life applications on phones, laptops, PCs, robots, and fast & lightweight server-side use-cases. > 8B MoE, 1.5B ac

arXiv:2605.27582v1 Announce Type: cross Abstract: Embodied navigation requires an agent to map language and visual observations to a stream of spatial actions that drive a real robot through environme

Utilidata Inc. today disclosed that it has raised 40 million in funding from Renown Capital Partners and Keyframe Capital. The cash infusion comes as an extension to a 60.3 million Series C round the

Artificial intelligence training data provider Human Archive Inc. today announced that it has raised 8.2 million in funding. Wing Venture Capital, NVP Capital, Y Combinator headlined the consortium th

AWS experienced margin expansion driven by growth in Anthropic-related services and an improved product mix from Amazon Bedrock, its generative AI platform, while competing cloud providers failed to a

arXiv:2605.26137v1 Announce Type: cross Abstract: While 3D generation is progressing rapidly, recent work has often focused on obtaining high-resolution assets, leaving user experience and deployabili

Nikkei Asia: Jensen Huang says Nvidia is spending 100B-150B annually on its Taiwan supply chain, up from 10B-15B four years ago, and will boost its Taiwan staff to 4,000 — TAIPEI — Nvidia is now spend

arXiv:2605.26636v1 Announce Type: cross Abstract: We introduce JetViT, a novel family of hybrid-architecture Vision Transformer (ViT) models that match the accuracy of state-of-the-art full-attention

arXiv:2512.01678v5 Announce Type: replace Abstract: Graph Neural Networks (GNNs) present a fundamental hardware challenge by fusing irregular, memory-bound graph traversals with regular, compute-inten

arXiv:2605.26430v1 Announce Type: new Abstract: Collaborative transport of objects via pushing by multiple robots has many applications, ranging from construction and warehouse environments to post di

arXiv:2605.26330v1 Announce Type: new Abstract: Small aerial robots are particularly well-suited for search and rescue in confined and hazardous environments due to their agility, low cost, and abilit

Nvidia CEO Jensen Huang announced the chip company plans to invest approximately 150 billion annually in Taiwan, terming it the 'epicentre' of the AI revolution . This represents a dramatic increase f

NVIDIA's Blackwell GB200 NVL72 architecture achieved the fastest-ever results on the STAC-AI benchmark for financial LLM inference, delivering up to 3.2x single-GPU performance improvements over the p

NVIDIA Dynamo Snapshot introduces ModelExpress for 7x faster startup via checkpoint restore and weight streaming with NVIDIA NVLink and NIXL , enabling efficient model initialization for inference wor

Nvidia has officially retired the NVIDIA Control Panel after 20 years of service, shifting all functionality to the newer Nvidia App. The legacy Control Panel is no longer included in default GeForce

arXiv:2602.01518v2 Announce Type: replace Abstract: Despite their importance in model sampling, efficient implementation of Top-k and Top-p algorithms for large vocabularies remains a significant chal

arXiv:2605.26632v1 Announce Type: new Abstract: Diffusion Transformers (DiT) achieve strong performance in image generation but incur substantial inference costs. While prior work has reduced this cos

arXiv:2605.27276v1 Announce Type: new Abstract: Humans are the bottleneck in building and improving AI. Both the models and the agents that wrap them are written, tuned, and corrected by people. The l

arXiv:2605.26509v1 Announce Type: new Abstract: Gaussian processes (GPs) provide a principled Bayesian framework for uncertainty estimation, but their computational complexity severely limits scalabil

Bloomberg: Sources: Taiwan suspects three people smuggled at least one Nvidia chip shipment to China after first exporting them to Japan; Taiwan detained them last week — Taiwan prosecutors suspect th

Tensormesh Inc. has hit upon a way to make artificial intelligence inference more efficient by eliminating the need for redundant computations, and its technology is so convincing that several of AI i

Enterprise GPU access is expanding as Advanced Micro Devices Inc. lowers barriers with the air-cooled Instinct MI350P GPU and a ready-to-run software stack designed for standard enterprise servers. Th

arXiv:2605.26720v1 Announce Type: new Abstract: Large language models (LLMs) have shown strong empirical gains as self-evolving agents for CUDA kernel generation, driven by feedback-conditioned planni

We're open-sourcing the Unigram tokenizer we rebuilt to reduce CPU utilization by 5-6x. Small rerankers and embedders run in single-digit milliseconds on GPU, making CPU tokenization a meaningful shar

NVIDIA DLSS 4.5 introduces Dynamic Multi Frame Generation and a second-generation transformer model for Super Resolution to enhance game image quality and frame rates for Unreal Engine 5. The new Tens

arXiv:2605.26118v1 Announce Type: cross Abstract: Porting deep learning algorithms to new hardware accelerators requires developers to repeatedly apply the same low-level optimizations -- quantization