Search: “nvidia-developer”

DFlash is an open source block diffusion model for speculative decoding that significantly accelerates LLM inference on NVIDIA Blackwell GPUs by drafting entire token blocks in parallel and verifying

NVIDIA Alpamayo 2 Super is a publicly available 34‑billion‑parameter vision–language–action model that merges a 32‑B Cosmos 3 Super Reasoner with a 2‑B action‑expert diffusion network. It produces uni

Meta released Muse Glimmer, a 30‑billion‑parameter dense language model with a context window exceeding 120 K tokens, designed for local, long‑running agentic AI workloads. The model is optimized to r

Developers can build with DeepSeek V4 through NVIDIA GPU-accelerated endpoints on build.nvidia.com, with hosted endpoints providing a fast way to prototype before moving to self-hosted deployment. Dee

The NVIDIA GB300 NVL72 platform set a world record by delivering 1,648 TFLOPs per GPU during the pre‑training of DeepSeek‑V3 671B, a mixture‑of‑experts (MoE) model. This achievement leveraged fifth‑ge

Alibaba released the open‑weights Qwen3.8‑2.4T‑A95B (Qwen3.8‑Max), a fine‑grained mixture‑of‑experts model with 2.4 trillion parameters, hybrid full‑ and linear‑attention, a one‑million‑token context

NVIDIA Alpamayo 2 Super is a publicly available 34‑billion‑parameter vision–language–action model that merges a 32‑B Cosmos 3 Super Reasoner with a 2‑B action‑expert diffusion network. It produces uni

World Action Models (WAMs) use video-based world modeling instead of vision‑language backbones, giving robots a learned physics engine that supports zero‑shot transfer to new tasks, environments, and

Meta released Muse Glimmer, a 30‑billion‑parameter dense language model with a context window exceeding 120 K tokens, designed for local, long‑running agentic AI workloads. The model is optimized to r

NVIDIA NeMo Switchyard is a routing platform that directs AI agent workloads to the most suitable specialized or frontier model for each step of a task, balancing performance, cost, and latency. It of

NVIDIA Nemotron 3.5 Lightning is an open‑source mixture‑of‑experts language model totaling 30 B parameters with only 3 B active during inference, designed to serve high‑volume, low‑latency execution f

NVIDIA JetPack 7.2.1 adds agentic video skills with the unified jetson‑videosdk, allowing programmable, device-aware video workflows that link developer intent to live device discovery and performance

Alibaba released the open‑weights Qwen3.8‑2.4T‑A95B (Qwen3.8‑Max), a fine‑grained mixture‑of‑experts model with 2.4 trillion parameters, hybrid full‑ and linear‑attention, a one‑million‑token context

A full‑stack observability framework for NVIDIA AI factories links telemetry from compute, networking, storage, orchestration and application layers using specialized tools (DCGM, NVSM, UFM, NetQ, NMX