October 21, 2026OverviewIn this episode of the AI-specialized technology podcast Latent Space, Evan Conrad, co-founder and ...

Nvidia plans to invest hundreds of millions in Firmus Grid, an Australian AI cloud firm that abandoned its $30B IPO after public market rejection.

TorchTPU is the native PyTorch backend for Google TPUs, now open source. The engineering milestone from April reaches the ...

Intel's new GPU D-series packs 480GB of memory for AI inference. Here's how it stacks up against Nvidia and AMD in 2026.

Intel officially named Crescent Island the GPU D-Series on Oct 12, 2026: 1.3 PFLOPS FP8, 1.5TB/s bandwidth, 480GB LPDDR5X at ...

Introduction: Three names, one questionCPU, GPU, TPU. These three terms, which always come up when talking about AI, are ...

Attention-FFN disaggregation hit commercial silicon at OCP Global Summit 2026: d-Matrix Squadron routed AI inference expert ...

Beyond traditional model scaling paradigms, quantization follows a distinct, data-driven scaling law that systematically maps ...

With the recent release of the Muse family of foundational models, we want to take you inside the infrastructure that has ...

Fabric.AI, Inc. (Nasdaq: FABC) is a fabless semiconductor company developing MicroLED-based optical interconnects designed to address the data-movement bottleneck in AI data centers. Its Neural I/o™ ...

Researchers at SLAC National Accelerator Laboratory have compressed a neural network by 98.8% and deployed it on an FPGA chip ...

In a normal year, back when Nvidia was a normal company, it would have been enough for RTX Spark laptops to “just work.” ...