NVIDIA AI News - Page 21 of 28

An AI factory packs GPUs, CPUs, switches, DPUs, and SuperNICs into a single operating system that also has to juggle schedulers, orchestration,...

Your next machine learning experiment will likely end up in a timestamped folder, alongside a notebook full of untested code and a fading memory of...

NVIDIA used IFA 2026 in Berlin to push local AI further into consumer hardware, pairing up with Microsoft and a stack of software partners to make...

Agility Robotics wants its humanoid robot on the same warehouse floor as human workers, no cage required.

NVIDIA has put out NemotronLabs VoiceChat 11B, an open speech-to-speech model built to handle live conversation without the usual pipeline of...

King Charles gathered some of the most powerful people in artificial intelligence at Dumfries House on Thursday, an 18th-century estate in Ayrshire...

The telecom industry dreams of networks that run themselves. It’s a powerful fantasy: AI predicting a cell tower’s failure before it happens,...

Some AI developers are bored with the present. They made a language model that thinks the world ended nearly a century ago.

Hugging Face confirmed this month that a fully autonomous AI system broke into its systems, an incident that drew gasps across the security world.

Claude Code talks to the Anthropic Messages API. Codex CLI talks to OpenAI's format.

OpenAI's own safety team labeled GPT-5 high-risk last summer, worried the model could walk someone with no scientific background through building a...

For six days, nobody on OpenRouter knew who built Ox Alpha. The model showed up free, unlabeled, and somehow good enough that hobbyists and indie...

Search has been lying to you. It tells you where your files are but has nothing to say about what's in them.

Anthropic put a number on its worry last month: GLM-5.3, the model Chinese company Z.ai released in August, poses what the company's report calls a...

A GB200 cluster running DeepSeek-V3 with no optimization managed just 103 TFLOPS per GPU, and 84% of that time went to GPUs talking to each other...

The 9.6 GB download is a promise. A 128K context window, a 4-bit quantized model, and the raw power of Gemma 4 sitting right on your NVIDIA RTX 2000...

The ceiling for on-device AI has always been a hardware problem. Training models with over 70 billion parameters?

Nvidia’s real business is selling shovels. Now it’s buying the whole mine. The company is committing $26 billion to develop open-weight AI models.

Google's latest AI chip, the Ironwood TPU, isn't just four times faster. The real headline is the staggering, tens-of-billions-dollar check the...

Seventy-five billion euros is not a bet. It’s a declaration, and SoftBank is building its fortress on a decommissioned military site in Bosquel,...