GTC 2026 - Page 9
Stories from GTC 2026 come first, newest first, then stories from connected topics (Jensen Huang, NVIDIA Vera Rubin, Nvidia, AI Chips & Compute, AI), each labelled with where it comes from. This view is not indexed by search engines.
Latest
Rambus Reports Record Q3 Revenue of $218.3m, Up 22% Year on Year
Rambus posted record Q3 2026 revenue of $218.3m, up 22%, as memory interface chip sales hit a record. It guides Q4 revenue to $227m-$233m.
Bedrock Robotics Raises $600m at $4.1bn to Scale Driverless Diggers
Bedrock Robotics, founded by ex-Waymo engineers, has raised $600m at a $4.1bn valuation after its autonomous excavators began paid work in August.
Unconventional AI to Buy NeuReality as It Chases a 1,000x Chip Goal
Naveen Rao's Unconventional AI has agreed to buy Israeli inference chipmaker NeuReality, bringing in about 40 engineers. No price was given.
SWI Compute Plans 30MW UK AI Factory, Orders About $1bn of NVIDIA Kit
SWI Compute says it has ordered about $1bn of NVIDIA systems and will open a 30MW UK AI factory by end-2027. It has not named the site or financing.
Cadence Says Maestro AI Models Lift Chip Design Quality Up to 30%
Cadence has launched Maestro, chip design AI models built partly on NVIDIA Nemotron that customers can run on-premises. Early access starts in Q4.
Meta Launches $35bn of Share Offerings and Plans a $60bn ATM Sale
Meta launched about $35bn of stock offerings on 26 October and plans a $60bn at-the-market sale of Class C shares from early 2027.
Reuters: Vietnam’s First Chip Fab Set to Miss 2027 Trial Target
Reuters sources say Viettel's chip fab, Vietnam's first, will probably miss its 2027 trial run, with no ASML deal yet. Viettel says it is on schedule.
SpaceX Says $40bn GPU Deal Is a Binding Lease Due to Close by Year-End
SpaceX says its $40bn GPU financing is a binding commitment to an Apollo-managed lease that is expected to close before year-end, a day after Elon Musk said the debt had closed.
NVIDIA’s Nemotron 3.5 Bolt E4B Packs an Agent Model Into About 4GB
NVIDIA's open Nemotron 3.5 Bolt E4B, built on Gemma 4 E4B, runs voice and device agents in about 4GB. Its benchmarks are NVIDIA's, and Qwen leads on some.
Block Says Its Dessa Voice Model Learned to Talk Without Any Text
Block's Dessa is a full-duplex speech-to-speech model trained only on human conversation. It runs on a MacBook Pro; early access is for safety researchers.
Latest on TSN
Explore the TSN Atlas: Every topic we cover, mapped. Click a bubble to dive in.