European AI factories everywhere, still Nvidia
Nvidia’s European AI factory expansion also strengthens the quantum-classical stack.
China’s LineShine tops the TOP500, and it did it without GPUs
China reaches No. 1 with scale, not AI efficiency.
Arm Ethos-N78 scales NPU IP to 10 TOPS
Second-generation Scylla architecture ships in hundreds of millions of Android SoCs.
Blaize GSP streams AI graphs at the edge
A hardware scheduler replaces the memory bottleneck—16 TOPS at 7 W.
Lumai’s free-space bet on optical inference
Lumai’s head of product explains the company’s approach to optical AI compute.
GreenWaves GAP9 ships AI in earbuds at 50 mW
RISC-V/NE16 hybrid processor wins tier 1 hearables sockets.
Why SRAM chips are AI’s favorite
We’re seeing a new class of chip based on SRAM.
‘The History of Visual Magic in Computers’—A summary companion
Jon Peddie supplements his recent book with a chapter companion and discussion guide.
Asygn’s Colibry NPU tries to make edge AI live on scraps of power
Asygn wants Colibry to bring vision AI to the smallest devices.
The AI effect
People need to come first when we use AI.
Google fills out the middle with the Gemma 4 12B
Using just 16GB of RAM, it can run on a consumer laptop.
Intel aims Crescent Island at inference
LPDDR5X memory targets practical AI deployment.