2+ day, 20+ hour ago   (470+ words) How full-stack serving optimizations increase user capacity on a 4xB200 system at a concrete interactivity target Deploying a large language model is only the first step toward production-ready serving. Production teams also need to serve as many concurrent users as possible…...

1+ week, 1+ day ago   (1308+ words) In September 2026, NVIDIA announced it is leaning into native GPU programming in Rust. CUDA C++ and CUDA Python are mature, enterprise-grade toolchains, and NVIDIA will be growing and maturing CUDA Rust into 2027 and beyond The systems layer of AI spans…...

1+ week, 5+ day ago   (571+ words) Cutting through the noise starts with one deceptively simple question: What problem are you solving? Different use cases map to wildly different infrastructure footprints. At a high level, most inference workloads fall into one of these four buckets: After mapping…...

2+ week, 3+ day ago   (1465+ words) Navigation enables a robot to turn perception and motion into purposeful autonomy. Unlike locomotion, which produces stable movement, navigation must be used to continuously localize the robot, interpret changing surroundings, select a route, and avoid obstacles to reach a goal…...

2+ week, 5+ day ago   (1398+ words) For years, a Python developer who needed a GPU had two realistic choices: Learn NVIDIA CUDA C++ well enough to write an extension, set up a build toolchain, and maintain bindings back to Python, which most people never did; or…...

3+ week, 1+ day ago   (23+ words) AI factories are interconnected systems where fleet economics depend on how efficiently the entire stack converts power and capital into completed agent tasks....

3+ week, 2+ day ago   (708+ words) AI safety and security teams at NVIDIA explore where security controls belong as AI agents take on increasingly complex work. Recent NVIDIA research underscores the importance of the harness layer in the agent stack. Using Agentic Variation Operators (AVO), researchers…...

3+ week, 4+ day ago   (695+ words) In production, RecSys models are often served online to millions of users under strict service-level agreements (SLAs), where small increases in latency can impact user experience. Unlike LLM workloads that may tolerate autoregressive decoding latency, RecSys models must frequently retrieve…...

3+ week, 4+ day ago   (972+ words) NVIDIA SkillEvaluator is an open source tool for measuring how skills affect agent performance through static checks and real-world task runs with and without each skill. NVIDIA verified Skills are packaged, signed capability descriptors that tell an agent exactly what…...

3+ week, 5+ day ago   (1463+ words) Teams customize their models to hit their targets for latency, speed, memory, and compute. With the open NVIDIA Nemotron family of models, developers can find the right-sized model for their needs. The new Nemotron 3.5 Lightning NVFP4 checkpoint, for example, preserves accuracy…...