Tag · 274 stories
NVIDIA
Every story tagged NVIDIA on AI Chat Daily.
NVIDIA makes the chips that train virtually every frontier AI model. Our NVIDIA coverage tracks earnings, hardware launches (Blackwell, Rubin), software strategy (CUDA, NIM), and the company's deals with frontier labs and hyperscalers.
NVIDIA — page 12
NVIDIA's Nemotron 3 Nano Omni claims 9x throughput edge over rival open multimodal models
The 30B-A3B mixture-of-experts model fuses vision, audio and text encoders, with a 256K context window and a 1920×1080 native input resolution.
NVIDIA's Omniverse stack hits 99% sim-to-real accuracy at ABB Robotics
ABB, JLR, Tulip and Terex post double-digit gains as OpenUSD-based simulation moves from pilot to production floor.
SusHi Tech Tokyo 2026 lands TechCrunch as media partner with a Disrupt fast-track
The April 27–29 event at Tokyo Big Sight will send one SusHi Tech Challenge winner to Disrupt's Startup Battlefield 200 in San Francisco.
Thinking Machines Lab raids Meta for talent as it lands Google Cloud deal
Weiyao Wang, an 8-year Meta veteran, is the latest researcher to defect to the $12B startup, which now counts roughly 140 employees.
DeepSeek's V4 lands with 1M-token context and a pivot to Huawei chips
The open-source flagship matches GPT-5.4 and Claude-Opus-4.6 on benchmarks while pricing V4-Pro at $1.74 per million input tokens.
Google Bets on Full-Stack AI With New TPUs, Chrome Agent and Thinking Machines Deal
At Cloud Next, Google unveils training and inference chips, an agentic Chrome, and a multi-billion dollar compute pact with Mira Murati's lab.
GeForce NOW adds Xbox Game Pass and Ubisoft+ labels to cloud library
Nvidia rolls out in-app tags first teased at GDC, plus six new cloud titles this week led by Vampire Crawlers and a Thor skin for Marvel Rivals.
TSMC triples CoWoS packaging for Nvidia Blackwell through 2027
The Taiwanese foundry commits to a 230,000-wafer-per-year CoWoS line, locking in Nvidia's next three generations.
Nvidia puts a petaflop of AI on your desk
Grace Blackwell's GB10 Superchip, packaged into Project DIGITS, delivers training-class compute from a standard wall outlet.
Stay ahead of everyone in AI.
The tightly edited AI news email engineers, founders, and investors actually open. One email. Every weekday. Five minutes to finish.
The briefing read inside teams at