Back to Articles Community Article Published October 21, 2026 Upvote 8 +2 Oluwaseun Doherty seun-doherty Follow nvidia Oyindamola Omotuyi OOyindamola Follow nvidia Richard AntRi31 Follow nvidia Asier Arranz asiernvidia…
Microduck is a newly released affordable open-source bipedal robot by Pollen Robotics and Hugging Face.
Models specialized for a domain or a task by fine-tuning general open-source models, such as Qwen and Gemma, inherit a vocabulary built for many domains and general applications.
In an essay grader we run in production, that vocabulary split the text the model writes most into many tokens.
Back to Articles Enterprise Article Published October 21, 2026 Upvote 1 Josh Longenecker jlongen-aws Follow amazon Many ML practitioners have some intuition about how to make models run fast, but fewer know how to analy…
In Part 1 of this series, you will learn how to systematically analyze inference performance and identify where the time is spent.
Back to Articles Enterprise Article Published October 21, 2026 Upvote 14 +8 George Saon gsaon Follow ibm-granite Brian Kingsbury bedkingsbury Follow ibm-granite Ron Hoory rhoory Follow ibm-granite Vishal Sunder vishalsu…
Together they cover the full range of enterprise speech work: transcription and translation with high accuracy, extremely fast transcription at the edge, and recordings where who spoke, and when, matter as much as what…
Claude Mods is an early form of the Attention Interface. It flips the agent harness paradigm from the harness facing the machine, to the harness facing the human.
For the last few weeks I’ve used Claude Mods to mold Claude Code to my AI engineering workflow.
Claude Code is great out of the box, but every engineering workflow is different.
Protein design is a search problem. Choose amino acids for an eight-residue protein sequence and more than 25 billion combinations are possible.
RFD4-Proteina co-designs protein sequences and all-atom structures using reward-guided inference-time search to focus compute on promising candidates.
The model combines explicit backbone coordinates with latent representation, removing the separate inverse-folding handoff required by prior approaches.
An AI agent can complete its assignment and still cross a security boundary. Recent incidents show that agents can circumvent security controls while using…
NVIDIA Sentry governs agent access requests, monitors runtime activity, and contains threats from an isolated hardware security layer in milliseconds.
NVIDIA Sentry monitors AI agents and enforces access policies from a hardware-isolated trust domain on BlueField-4 DPUs, independent of host software.
Autonomous agents can work for hours on complex tasks, accessing files, tools, the internet, and spawning other agents to collaborate with.
NVIDIA OpenShell provides an open-source runtime that enforces security and governance for agentic systems by treating agent workloads as untrusted and granting access only when needed.
The framework uses four components — Gateway, Supervisor, Sandbox, and Policy Prover — to manage sandbox lifecycles, mediate credential access, isolate workloads, and formally verify policy permissions.
Building an AI factory digital twin requires developers to bring site plans, equipment assets, and engineering requirements into a consistent model that teams…
The NVIDIA Omniverse DSX Blueprint provides reference digital twin assets, sample scripts, and workflows for developing AI factory digital twins on OpenUSD and NVIDIA Omniverse libraries.
An agent-assisted workflow combines DSX reference designs, architectural references, site plans, requirements documents, and geographic data into an editable OpenUSD campus model that teams can inspect and revise.
AI factories turn energy into intelligence. But before an AI workload can produce its first token, operators must bring an interconnected system of compute…
NVIDIA DSX POD is a standardized, factory-built reference design that pre-integrates rack-scale AI infrastructure into containers for validation before shipment.
The architecture aligns with the NVIDIA Vera Rubin platform and defines coordinated requirements across compute, networking, storage, power, liquid cooling, cabling, and mechanical design.
Proudly brought to you by SerpAPI: Today’s release covers a LOT of ground, but you can scrape all this into markdown for your agents !
We last covered the Inference Inflection in April , and historically most of the advances have been closed source decacorns from Fireworks to Baseten to Modal .
Listen now (44 mins) | Paying for outcomes, building with agents, and the last mile of enterprise AI adoption
Last year , Alex Lieberman and his cofounder joined us to explain Tenex’s unusual bet: pay AI engineers for the work they deliver, rather than the hours they spend at a keyboard.
If AI makes an engineer faster, an hourly rate can make the engineer less money for the same job.
Back to Articles Community Article Published October 21, 2026 Upvote 6 Lattice Admin LatticeAdmin Follow The Lattice 4D Dataset is an early, publicly available collection of real-world human-object interactions, showcas…
The dataset combines multiview RGB-D recordings, reconstructed geometry, human motion, derived egocentric views, structured world states, and LeRobot-compatible data to support robotics and physical AI research.
Back to Articles Enterprise + Article Published October 20, 2026 Upvote 4 Lukas Zbinden lzbinden Follow nvidia Javier Gamazo javirk1 Follow nvidia Nigel Nelson nigeln Follow nvidia Mostafa Toloui mtoloui Follow nvidia S…
Today, we take the next step and introduce Cosmos3-H-Dreams, bringing interactive generative simulation to the visual and procedural complexity of clinical surgery with our most convincing in-body imagery to date.
Back to Articles Community Article Published October 21, 2026 Upvote 6 Saira Qureshi sairaq Follow nvidia Aditi Gautam adgautam Follow nvidia Manikandan Ravikiran mravikirannvidia Follow nvidia Sena Ekiz neptunel1 Follo…
It combines diffusion modeling, curriculum learning, and adaptive thresholding to identify unusual behavior in noisy, high-dimensional signals.
Back to Articles Community Article Published October 14, 2026 Upvote 13 +7 VIDRAFT_LAB SeaWolf-AI Follow FINAL-Bench Most neural processing units (NPUs) in the market are inference accelerators.
The usual assumption is that you can serve on them, maybe quantize for them, but you cannot train on them, and certainly not pretrain a model from scratch.
Using GPT-4.1, an older, non-reasoning model, a basic customer-service agent reaches 67.5% Pass^4 on τ²-bench retail — meaning it solves the task on all four independent attempts for 67.5% of tasks.
It scores above every retail entry in our October 12, 2026 leaderboard snapshot.
Back to Articles Community Article Published October 20, 2026 Upvote 7 +1 Emma Scharfmann EmmaScharfmann Follow Aritra Roy Gosthipaty ariG23498 Follow Kashif Rasul kashif Follow Applying the downscaler we trained on AIF…
Ensemble mean of the forecast over France, valid Wednesday 13 August 2025 at 12:00 UTC (+60 h from initialisation).
