Knowledge catalogue
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
arXiv:2601.05289v2 Announce Type: replace-cross Abstract: The high-dimensional complex nature of detectors makes fast calorimeter simulations a prime application for modern generative machine learning
After seeing these tweets, I decided to try it out on my own old Ubuntu computer with RTX 1070 GPU (the one that I just upgraded from 16.04 all the way to 24.04 the other day). Asked Codex on my Mac t
arXiv:2605.25092v1 Announce Type: cross Abstract: Long-term conversational memory is a retrieval workload classical IR was not built for: the index grows during the query stream, query types shift int
This Databricks blog post likely examines the prerequisites and preparedness required for telecommunications companies to successfully implement and deploy AI solutions. It probable covers organizatio
Agents are turning up the heat for enterprises, turning air-cooled AI infrastructure into a boardroom priority. While GPUs have largely dominated the AI conversation, CPUs are increasingly coming into
arXiv:2605.24684v1 Announce Type: cross Abstract: Multimodal Attributed Graph Learning (MAGL) integrates intrinsic node attributes with structural topology via graph aggregation. However, as pretraine
In this post you'll learn how to build a multi-agent campaign review system that demonstrates parallel reasoning, context persistence, and traceable execution paths using an integrated architecture th
This NVIDIA Developer article guides developers on creating optimized GPU kernels using C++ with NVIDIA's CUDA Tile technology, which enables efficient parallel computation on graphics processors. The
Jonathan Vanian / CNBC: Dropbox founder Drew Houston is stepping down as CEO after 19 years to become executive chairman, replaced by Ashraf Alkarmi, who is SVP and GM of Dropbox Core — Drew Houston f
arXiv:2605.24930v1 Announce Type: new Abstract: Transformer-based LLMs achieve strong results on many language tasks; however, long inputs remain challenging because context windows are finite, and pr
I have been very impressed by @SemiAnalysis_ . I think of myself as a wide ranging systems engineer, looking for value at every level from the chip specs to the user interface, but SA exposes me to ad
arXiv:2605.24168v1 Announce Type: new Abstract: Sparsity has long been a central theme in LLM efficiency, but its role in context processing remains unresolved. As LLM workloads shift toward longer co
Michael Larabel / Phoronix: Initial benchmarks show Nvidia's Vera CPU, which features 88 in-house-designed Olympus cores, packs a heavy-hitting punch, beating Intel's and AMD's x86_64 CPUs — NVIDIA's
This article examines the emerging 800V direct current (800VDC) electrical architecture in electric vehicles, which represents a significant advancement over traditional 400V systems by enabling faste
Inside the 800VDC Revolution – Part 1 Four-Phase 800VDC Transition, Power Rack Economics, SST, Equipment Content/MW Build, Supplier Implications https://newsletter.semianalysis.com/p/inside-the-800vdc
arXiv:2602.05052v3 Announce Type: replace Abstract: We present a unified algorithmic framework for the numerical solution, constrained optimization, and physics-informed learning of PDEs with a variat
arXiv:2507.03159v2 Announce Type: replace Abstract: We present exttt{MathOptAI.jl}, an open-source Julia library for embedding trained machine learning predictors into a JuMP model. exttt{MathOptAI.jl
Tom Warren / The Verge: Nvidia officially retires its GeForce Control Panel app after 20 years, following the porting of all of its major features to the Nvidia app — Nvidia has now ported across all
Artificial intelligence inference routing startup OpenRouter Inc. today announced it raised 113 million in new funding led by CapitalG, Alphabet Inc.’s independent growth fund, to push the envelope fo
arXiv:2605.25124v1 Announce Type: new Abstract: The Gini Multidimensional Scaling (Gini MDS) framework extends the Euclidean multidimensional scaling. We introduce a Gini pseudo-distance based on valu
arXiv:2605.25346v1 Announce Type: cross Abstract: Neural network (NN) dynamics models and control policies achieve strong performance in robotics, but providing sound guarantees under uncertainty rema
arXiv:2605.26064v1 Announce Type: cross Abstract: We present Paris 2.0, the first video generation model pre-trained through decentralized computation. Its training recipe builds upon Paris 1.0 (arXiv
arXiv:2602.07518v2 Announce Type: replace-cross Abstract: Kolmogorov-Arnold Networks (KANs) shift neural computation from linear layers to learnable nonlinear edge functions, but implementing these no
Rambus Inc. today announced a complete DDR5 9600 client memory module chipset designed to push PC memory speeds to 9,600 megatransfers per second, targeting the bandwidth and capacity demands of agent
recommended reading. i too am very done with people anthropomorphizing a bunch of matrices on a GPU cluster, especially if the same people do not give two fucks about actual human beings. More musings
The NVIDIA RTX PRO 4500 accelerates genomic sequencing and drug discovery by slashing processing times. This professional desktop GPU features 32GB of ultra-fast GDDR7 memory and Blackwell architectur
arXiv:2605.24016v1 Announce Type: cross Abstract: Diffusion inference remains costly for edge deployment, yet existing accelerators focus almost exclusively on score networks because standard drift is
We aren’t going to do this again so quickly, are we? Rising demand results in higher costs. Higher costs result in lower demand. It is almost like some sort of equilibrium is being achieved. But there
Anthropic is really a mixed bag; they rightly support moderate regulation, but they invite wildly anthropomorphic language about LLMs, and steal IP just like their peers. Anthropic are the bad guys. W
Chinese electronics giant Huawei Technologies Co. Ltd. today unveiled a new chip design framework that it says will help it to close the gap in the semiconductor industry with global leaders like Taiw
arXiv:2605.23235v1 Announce Type: new Abstract: Globalization and multiculturalism continue to produce increasingly diverse speech varieties. Yet current spoken dialogue systems frequently fail on und
As organizations push to convert AI ambition into defensible business results, many are discovering that infrastructure alone is not enough. Instead, the real challenge lies in reconciling decades of