NVIDIA released the Nemotron 3.5 Lightning NVFP4 checkpoint, compressed from 66 GB to 22 GB via 4-bit quantization, achieving up to 4x faster throughput while preserving accuracy. The model was develo…
Alibaba released the open weights for Qwen3.8-2.4T-A95B (Qwen3.8-Max), a 2.4T-parameter mixture-of-experts model with 95B activated parameters per token, designed for long-context reasoning and agenti…
NVIDIA published a guide for selecting full-stack observability tools for AI factories, outlining a decision framework that maps components such as GPUs, InfiniBand fabrics, and cluster schedulers to …
NVIDIA JetPack 7.2.1 adds support for PyNvVideoCodec 2.2 on Jetson Thor, enabling Python developers to use hardware-accelerated video encoding and decoding with GPU-resident memory. The release also i…
NVIDIA released Nemotron 3.5 Lightning, an open 30B mixture-of-experts model with 3B active parameters, optimized for high-volume, low-latency execution in long-running AI agents. The model, part of t…
NVIDIA introduced NeMo Switchyard, a library that enables developers to route AI agent workloads across multiple models based on task requirements, cost, and latency, improving efficiency and accuracy…
Meta released Muse Glimmer, a 30B open-weight dense model with a 120K+ context window, optimized for local AI agentic workflows on NVIDIA platforms, delivering over 20K tokens/sec on a single GPU. The…
NVIDIA released Alpamayo 2 Super, an open 34-billion-parameter reasoning vision-language-action model for autonomous vehicle development, combining a 32-billion-parameter Cosmos 3 Super Reasoner with …
NVIDIA Corporation announced benchmark results showing its Vera CPU, featured in the NVIDIA Vera BlueField-4 STX Storage Processor, outperforms x86 CPUs in encryption, compression, integrity checking,…
NVIDIA has published a technical post analyzing how group size, head dimension, and sequence length affect dense attention performance in long-context inference, offering a co-design checklist to help…
NVIDIA released Video Codec SDK 13.1, adding AV1 Hierarchical Reference Mode with up to 31 B-frames, per-macroblock decode statistics for H.264 and HEVC, frame-accurate seek, and application-allocated…
NVIDIA released nvmath-python v1.0, a Python library that provides access to NVIDIA CUDA-X math libraries for high-performance core math operations on CPU, single GPU, or distributed multi-GPU, multi-…
NVIDIA's AI Red Team found that the most common failure mode in AI agent deployments is a lack of access control, with many agents holding user credentials accessible to any authorized network user. T…
NVIDIA reports that two AI computing clusters built from identical H100, GB200 NVL72, or GB300 NVL72 systems can deliver 8% to 12% differences in training throughput due to configuration choices in ke…
NVIDIA has published a tutorial for self-hosting a validated AI coding assistant using StarCoder2-7B NIM and NeMo Guardrails, enabling code completions on local GPUs with policy enforcement, hallucina…
NVIDIA released Ising Calibration 1.5, an open-source vision language model for fully automated quantum processor calibration, which is 11.4% smaller at BF16 precision and 86.68% better than its prede…
NVIDIA Labs released the Object-Oriented Agents (NOOA) research preview, an open-source Python framework that treats AI agents as single Python classes to improve performance and developer workflow. T…
NVIDIA's Nemotron 3 Ultra large language model, when paired with the ACE-RTL agent, achieves a 100% pass rate on the Comprehensive Verilog Design Problems (CVDP) benchmark for agentic RTL coding tasks…
Applied Materials and NVIDIA are collaborating on an end-to-end digital development model that combines Applied Materials' materials engineering and semiconductor manufacturing expertise with NVIDIA C…
NVIDIA OptiX Toolkit (OTK) provides debugging facilities for GPU ray tracing applications, including consistent checking of OptiX and CUDA API error codes and targeted device-side debug printing. OTK,…