Knowledge catalogue
arXiv:2607.29397v1 Announce Type: new Abstract: Deploying large language models in realistic server environments poses challenges, as the system needs to provide high-quality responses with low latenc
The Inference Engineering Masterclass: 10x faster models, quantization, speculative decoding, Rubin, & self-optimizing AI https://www.latent.space/p/inference-eng @Baseten @philipkiely and @waterloo_i
The math still ain’t mathing. Recently we estimated global (ex China) AI revenues of around 200bn annualised, based on four different sources. Just come across a fifth, based on Nvidia inference sales
arXiv:2607.29678v1 Announce Type: new Abstract: LLM serving systems cache prompt KV state, yet most front ends still re-tokenize the full request text on every call. The cost lands on coding agents, w
arXiv:2607.28633v1 Announce Type: cross Abstract: Disaggregated LLM inference creates a datacenter networking problem that no existing system solves correctly. When prefill and decode run on separate
Hermes Agent is now dramatically more efficient, especially for smaller/weaker/local models! With the help of @nvidia's Nemo Relay and several other strategies Hermes was able to identify a ton of opt
Hugging Face CEO Clément Delague says “AI is actually an opportunity to fix a lot of the cybersecurity problems” because his company used Nvidia’s version of a Chinese open model to defend itself agai
I deployed K3 on 32 H100s at work a couple of weeks ago and then got annoyed that there was no way to poke at it on my own machine. So I wrote an inference engine for it in C99. Nothing clever going o
The funny thing about Anthropic and OpenAI people saying they want to slow down AI progress is that this is what Anti Trust mechanisms were built for. It's illegal to collude and slow down AI progress
Watching @ClementDelangue on @FaceTheNation discussing agentic hacking. “Preventing releases does not work; concentrating behind closed doors in just a few organizations doesnt work. What worked in th
Banning open source model, will 'remove capabilities for the defenders' When OpenAI's models breached Huggingface, a Chinese open model is what cleaned up the mess. Anthropic's Fable 5 refused, so Hug
Vivid depiction of a dark scenario for Nvidia. “Indeed, CEO Jensen Huang has been so aggressive in funding the AI ecosystem that the company is functioning almost like a bank; with annual free cash fl
arXiv:2607.28047v1 Announce Type: cross Abstract: Time-varying implicit neural representations (INRs) provide a compact representation of scientific volumes and, for modalities such as dynamic X-ray c
arXiv:2607.26661v1 Announce Type: new Abstract: Ascend C operator optimization is critical for NPU (Neural Processing Unit) inference performance but requires deep hardware expertise.While large langu
GPU utilization can read healthy while your queue backs up, and a new replica takes minutes to warm. Here's how to pick autoscaling metrics, tune scale-up/down windows, and budget for cold starts on d
The article shows that dense‑attention performance in long‑context inference is governed by group size (query heads per KV head), head dimension, and sequence length, with prefill being compute‑bound
arXiv:2607.27945v1 Announce Type: cross Abstract: Sequence models must decide what to write into memory and what to retain. In quantum and quantum-inspired sequence learning, nonlinear recurrent updat
arXiv:2607.26723v1 Announce Type: cross Abstract: Inversion-based watermarking is a promising approach to authenticate diffusion-generated images, yet practical use is bottlenecked by inversion that i
arXiv:2607.28416v1 Announce Type: new Abstract: Curved tactile fingertips for dexterous manipulation must resolve fine contact geometry, distinguish normal and tangential loads, and capture transient
arXiv:2607.26076v1 Announce Type: cross Abstract: Retrieval-augmented generation services over mutable enterprise documents repeatedly execute semantically equivalent analysis requests. Answer reuse c
arXiv:2607.26181v1 Announce Type: new Abstract: Functional verification dominates integrated circuit (IC) front-end engineering effort, and a single missed bug that escapes to silicon can trigger a co
Justine Moore announced on July 31, 2026 that Flux 3’s latest iteration excels at generating GPU‑based ASMR content. She confirmed that this capability is now available in an early preview on the Nous
Data center builder Nscale Global Holdings Ltd. today announced plans to acquire Anyscale Inc., a venture-backed provider of artificial intelligence software. The terms of the deal were not disclosed.
NVIDIA Video Codec SDK 13.1 adds AV1 hierarchical reference mode supporting up to 31 B‑frames and efficient iterative tuning that delivers significant bitrate savings in CQ and VBR modes. It enhances
arXiv:2607.28312v1 Announce Type: new Abstract: Streaming video understanding requires models to continuously retain useful visual evidence before future questions are known. Existing approaches prima
OPENAI IS GOING TO TAKE THIS ENTIRE MARKET DOWN WITH IT And you don't have to own a single share to get hurt. What I'm about to explain should worry anybody who thinks they're diversified: OpenAI is a
arXiv:2607.28144v1 Announce Type: cross Abstract: We present ReGenVC, an end-to-end generative video codec that compresses talking-head video to an ultra-low bitrate and decodes it in real time. The e
arXiv:2607.28627v1 Announce Type: new Abstract: Long visual context poses a challenge for vision-language models: performance degrades as the number of distractors grows, and processing all tokens at
arXiv:2607.27744v1 Announce Type: new Abstract: Modern recommendation models gain prediction quality by scaling feature-interaction and sequence modules, but production cost constraints cap how far sy
Security at the foundation requires openness at the foundation. As open-weight models become critical infrastructure for the next generation of software, the systems around them must be transparent, i
arXiv:2607.27494v1 Announce Type: new Abstract: Recent advances in robotics research have created a strong demand for high-performance simulators. Surgical robotics simulation faces unique challenges
Mackenzie Hawkins / Bloomberg: Sources: Moonshot has a computing power agreement with Alibaba for the use of ~20K Nvidia chips; some say the deal is for H200 chips, which Alibaba denies — Chinese AI c
arXiv:2607.28032v1 Announce Type: new Abstract: Creating photorealistic animatable head avatars from a single image remains a fundamental challenge in digital human synthesis. While recent 3D Gaussian
The first thing we learned building autoscaling for dedicated inference is that CPU-style metrics don't tell the whole picture. A GPU can read 60% busy while the engine's queue is already backing up,
Together AI gives developers a high-throughput production path for running Inkling-Small on NVIDIA Accelerated Infrastructure across multimodal, coding, and agentic workloads. Start building: https://
arXiv:2607.28263v1 Announce Type: new Abstract: Transformer depth is not used uniformly: lower and middle layers build semantic representations, while upper layers increasingly specialize them for pre
arXiv:2607.26829v1 Announce Type: new Abstract: Many high-performing volumetric segmentation models maintain dense multi-scale feature maps, leading to high activation memory and inference cost. We pr
Back to school means balancing assignments, deadlines and downtime. GeForce NOW makes it easy to have it all. With cloud gaming, everyday laptops used for class can also become GeForce RTX-powered gam
Can AI agents conduct open-ended AI research? Most evaluations of agents conducting AI research focus on narrow, verifiable tasks. But AI research is often open ended. Researchers pick hypotheses, dec
Osmosis_AI claims that an open‑source model can rival a foundation model by leveraging reinforcement learning techniques. To demonstrate this, they will use the Y Combinator‑dedicated GPU cluster on T
NVIDIA’s AI Red Team found common failure modes in enterprise AI agents: weak access controls, unrestricted code execution via tools, unprotected network egress, and exposure of plaintext secrets. The
arXiv:2607.26646v1 Announce Type: new Abstract: We address the problem of reconstructing a high-fidelity, freely navigable 3D scene from a single 360^irc panorama, without per-scene optimization or mu
Hey everyone, I recently open-sourced a new Python package called ganfs (Generative Adversarial Network Feature Selection), and I wanted to share it with the community. The Problem: Selecting the best
arXiv:2607.26238v1 Announce Type: new Abstract: We investigate lightweight raptor-species classification for real-time edge deployment in wind-turbine collision mitigation. Using DINOv2-L (304M parame
arXiv:2607.26855v1 Announce Type: new Abstract: Many scientific fields rely on standard benchmarks and shared platforms to improve review and reproducibility, but autonomous systems research still lac
NVIDIA’s Exemplar Cloud study shows that identical H100‑based clusters can yield 8–12 % lower training throughput when configuration gaps—at the kernel, hypervisor, BIOS or NCCL levels—prevent reachin
arXiv:2510.12346v2 Announce Type: replace Abstract: Recently, biped robot walking technology has been significantly developed, mainly in the context of a bland walking scheme. To emulate human walking
Enterprise AI infrastructure providers are turning to multi-tenancy paired with upstream data protection to convert idle GPU capacity into secure, token-metered services that enterprises will actually
John Thornhill / Financial Times: Q&A with CuspAI's Max Welling on its AI Materials Foundry, partnerships with Nvidia and others, Geoff Hinton and Yann LeCun joining its advisory board, and more — The
NVIDIA **nvmath‑python v1.0** is a Python library that wraps CUDA‑X and NVPL math libraries (cuFFT, cuBLASLt, cuDSS, cuSPARSE, cuTENSOR, cuBLASMp) to give NumPy, CuPy, and PyTorch users GPU‑accelerate
Are there any benchmarks on these 4 bit quants, like how Artificial Analysis runs a slew of various benchmarks? If not, how can I run one (5x over for consistency) on them? I'm also very interested in
arXiv:2607.26694v1 Announce Type: new Abstract: We present Visko Orbis 1.0, a Live Model for real-time, interactive long-video generation. Users can change the prompt at any moment during generation,
Artificial intelligence has propelled the cybersecurity world into a new phase, driven by startlingly advanced autonomous attacks and an urgent need to adopt technology to defend against them. This we
agreed. which is part of why coating the world in data centers is a profound mistake. AI will get so ridiculously efficient that we will look back at GPU clusters the way we now look at these first ro
It's a drop-in. One program_id field and it plugs into your existing engine configs such as KV offloading + speculative decoding. Format-agnostic by design, with OpenAI chat completions supported toda
arXiv:2607.26016v1 Announce Type: cross Abstract: Recently, photonic transformer accelerators (PTAs) have successfully achieved significant speedup and energy efficiency improvements over electronic a
arXiv:2607.25967v1 Announce Type: new Abstract: Singular Value Decomposition (SVD) underlies matrix factorisation tasks across computational imaging, with medical applications increasingly demanding r
Super interesting new work from NVIDIA. (bookmark it) They suggest building agents as Python objects. Very cool idea and I think it could a lot with agent reliability. More below: Agent development to
arXiv:2505.14366v2 Announce Type: replace Abstract: We present a conceptual framework for training Vision-Language Models (VLMs) to perform Visual Perspective Taking (VPT), a core capability for embod
arXiv:2607.22598v1 Announce Type: cross Abstract: Educational chatbots powered by large language models (LLMs) show promising effects on learning outcomes, yet most systems delegate pedagogical decisi