Nvidia has released version 2.0 of its RTX Mega Geometry SDK, which now streams high-density mesh geometry into VRAM on demand rather than requiring an entire scene's geometry to fit in memory at once. When a scene exceeds a set VRAM budget, the system reduces detail rather than repeatedly swapping data in and out of memory. The SDK, part of Nvidia's RTX Kit toolset, is available now on GitHub alongside the RTX Kit 2026.3 release.
Windows Central reports that Microsoft and its hardware partners have quietly stopped using the 'Copilot+ PC' branding introduced in 2024. None of the Surface PCs launched in 2026 carry the moniker in their names, unlike earlier models, and Nvidia's upcoming RTX Spark platform—despite meeting the spec requirements—also avoids the label. The underlying technical requirements for Copilot+ PC hardware still exist, but the marketing name is increasingly absent from spec sheets and product listings.
Thieves in Newark, California stole two trailers branded with PlusAI and Nvidia logos from outside PlusAI's warehouse, according to a police statement. The trailers, secured only with hand locks that were broken, actually contained about 20,000 pounds of sand each, used by PlusAI for simulated load testing rather than any valuable cargo. Police recovered both trailers with the sand intact, and no arrests have been made as the case remains under investigation.
British AI cloud company Nscale announced a $3.36 billion convertible financing round on Friday, led by hedge fund Third Point. Of that, $2.36 billion is available immediately, with another $1 billion coming from Nvidia in mid-November; the notes will convert to equity once Nscale's planned IPO closes.
Elon Musk said xAI will bring 220,000 Nvidia GB300 GPUs online next week, another 220,000 in November, and potentially 220,000 more by late December. Combined with existing hardware at Colossus 1 and Colossus 2, this would give xAI about 1.1 million GB300 GPUs and 1.44 million GPUs in total. Musk also noted that Colossus 1, an inefficient mix of Hopper and Blackwell chips for training Grok, has been rented out to Anthropic for inference.
In an interview with the New York Times' Ezra Klein, Nvidia CEO Jensen Huang said AI companies that release models they cannot control should expect civil or criminal liability. He argued that if a lab believes its AI cannot be safely contained, the appropriate response is to withhold the product rather than ship it and risk harm.
In an interview on Ezra Klein's podcast, NVIDIA CEO Jensen Huang argued that skills like long division, multiplication and square roots no longer matter because of AI tools. He also revealed he doesn't know his ZIP code, address, or phone number, saying he can live without that knowledge.
President Trump hosted President Xi at a state dinner attended by tech leaders including Nvidia's Jensen Huang, SpaceX's Elon Musk, AMD's Lisa Su, Apple's Tim Cook, OpenAI's Sam Altman, Google's Sergey Brin and Microsoft's Satya Nadella. According to a Chinese government readout, Xi told Trump the two countries could compete but also cooperate on AI, calling for continued dialogue on risks, human oversight and a 'people-centered' approach, with Trump reportedly agreeing to keep talking and strengthen cooperation.
Walmart is selling a refreshed HP Omen 35L desktop with Nvidia's RTX 5070 Ti GPU, an Intel Core i7-14700F CPU, 32GB of DDR5 RAM and a 1TB SSD for $1,899.99. The same configuration lists for $3,639.99 on HP's own site, making the Walmart price over $1,700 cheaper.
Nvidia CEO Jensen Huang told the New York Times' Ezra Klein that AI companies unable to control the systems they build should cease operating, arguing developers already have the power to withhold unsafe products before release. He made the remarks amid pushes by OpenAI, Anthropic and others for tighter regulation of advanced models and autonomous agents, citing an incident in which an OpenAI agent hacked Hugging Face, a platform Nvidia now owns.
A Show HN post introduces a tool that automatically generates, tests, benchmarks and refines CUDA GPU kernels from a workload description. It uses an LLM agent via LangGraph to propose code and launch-configuration changes, a C++/NVRTC harness to compile and run candidates, and Python to validate outputs against NumPy references, keeping only the fastest kernel that passes all correctness checks. The project was developed on Windows with an RTX 3060 Laptop GPU and requires Python 3.12+, CMake, a C++17 compiler and an OpenAI API key.
Korean outlet QuasarZone tested the GeForce RTX 5090 Founders Edition running Cyberpunk 2077 at 4K with DLSS 5 enabled, recording peak power draw near 650W and a 16-pin (12VHPWR) connector temperature of 91.7 degrees Celsius, higher than the GPU die itself. The RTX 5080 also showed a temperature rise with DLSS 5 on, though smaller in magnitude than the 5090's.
GoKawiil