Submissions from nvidia.com

		Custom FP4 CUDA Kernel – 129 Tflops on DGX Spark with Pre-Quantized Weight Cache (nvidia.com)
		1 point by vkaufmann 3 hours ago \| past \| 1 comment
		Ppisp: Cleaner Representations of Gaussian Splats and NeRFs (nvidia.com)
		7 points by totalview 6 days ago \| past \| discuss
		Python All the Way Down: Speed-of-Light CUDA Without Leaving Python (nvidia.com)
		2 points by pjmlp 6 days ago \| past \| discuss
		Nvidia PersonaPlex: Natural Conversational AI with Any Role and Voice (nvidia.com)
		3 points by matco11 9 days ago \| past \| discuss
		Run OpenClaw for Free on GeForce RTX and Nvidia RTX GPUs and DGX Spark (nvidia.com)
		2 points by teleforce 10 days ago \| past \| discuss
		Inference Providers Cut AI Costs by 10x with Open Source Models on Blackwell (nvidia.com)
		4 points by gmays 10 days ago \| past \| discuss
		Automating Inference Optimizations with NVIDIA TensorRT LLM AutoDeploy (nvidia.com)
		1 point by matt_d 13 days ago \| past \| discuss
		Quantization-Aware Distillation for NVFP4 Inference Accuracy Recovery [pdf] (nvidia.com)
		2 points by gmays 17 days ago \| past
		Windows 11 update KB5074109 reportedly reduces gaming performance (nvidia.com)
		5 points by LopRabbit 18 days ago \| past
		Explaining Tokens – The Language and Currency of AI (nvidia.com)
		1 point by brudgers 18 days ago \| past
		Fusing communication and compute with new API and copy engine collective in NCCL (nvidia.com)
		1 point by teleforce 19 days ago \| past
		NVIDIA Transformer Engine (nvidia.com)
		2 points by teleforce 20 days ago \| past
		Meta's Grand Teton Brings Nvidia Hopper to Its Data Centers (nvidia.com)
		1 point by onurkanbkrc 22 days ago \| past
		Nvidia: Quantization-Aware Distillation for NVFP4 Inference Accuracy Recovery [pdf] (nvidia.com)
		4 points by tosh 26 days ago \| past
		Nvidia PersonaPlex: Natural Conversational AI with Any Role and Voice (nvidia.com)
		3 points by smusamashah 26 days ago \| past
		GeForce Now Brings GeForce RTX Gaming to Linux PCs (nvidia.com)
		27 points by pjmlp 26 days ago \| past \| 11 comments
		Think Smart About Sparse Compute: LatentMoE for Higher Accuracy per Flop, Param (nvidia.com)
		2 points by buildbot 28 days ago \| past
		Nvidia PersonaPlex: Natural Conversational AI with Any Role and Voice (nvidia.com)
		3 points by virgildotcodes 28 days ago \| past \| 1 comment
		Nvidia Earth-2 (nvidia.com)
		4 points by meistro 28 days ago \| past \| 1 comment
		Using Context as Training Data Unlocks Models That Learn at Test-Time (nvidia.com)
		1 point by santadays 29 days ago \| past
		NVIDIA Launches Earth-2 Family of Open Models and Tools for AI Weather (nvidia.com)
		3 points by geox 29 days ago \| past
		Nvidia PersonaPlex: Natural Conversational AI with Any Role and Voice (nvidia.com)
		2 points by wiradikusuma 31 days ago \| past
		Nvidia: Dynamic Memory Compression (nvidia.com)
		2 points by alecco 31 days ago \| past
		Nvidia PersonaPlex: natural conversation AI (nvidia.com)
		3 points by ricardobeat 33 days ago \| past
		How to Train an AI Agent for Command-Line Tasks with Synthetic Data and RL (nvidia.com)
		1 point by gmays 33 days ago \| past
		Nvidia: Natural Conversational AI with Any Role and Voice (nvidia.com)
		13 points by websku 35 days ago \| past \| 3 comments
		BioNeMo Platform Accelerate AI-Driven Drug Discovery (nvidia.com)
		2 points by gmays 38 days ago \| past
		Using Context as Training Data Unlocks Models That Learn at Test-Time (nvidia.com)
		1 point by krautburglar 42 days ago \| past
		Nvidia: Using Context as Training Data Unlocks Models That Learn at Test-Time (nvidia.com)
		6 points by ashvardanian 43 days ago \| past
		Nvidia Announces Alpamayo Open-Source AI Models to Accelerate Reasoning-Based AV (nvidia.com)
		2 points by lateforwork 45 days ago \| past
		More