Why the first GPU financiers are turning to inference chips in a $400 million deal
A $400 million chip-backed loan points to the next wave of AI infrastructure deals.
Get smart on it
An AI inference startup has secured a $400 million loan using inference-specific chips as collateral, marking what appears to be the first time such chips have been used this way. Inference chips are designed to run already-trained AI models quickly and efficiently, unlike the more expensive chips used to build models initially. This financing reflects growing market interest in cheaper alternatives to running the newest large language models, with investors betting that open source models and inference infrastructure will become increasingly important. The deal also signals a shift away from Nvidia's dominance, as investors now see opportunity in companies building around alternative chip makers rather than the traditional GPU ecosystem.
Related news
Designing High-Performance GPU Kernels with TileLang: Tensor-Core GEMM, Fused Softmax, FlashAttention, and Autotuning
In this tutorial, we explore TileLang as a high-level Python domain-specific language for designing and compiling performance-oriented GPU kernels through TVM. We begin by validating the CUDA environment and establishing reusable benchmarking and numerical-verification utilities, then progressively implement vector addition, tiled tensor-core matrix multiplication, schedule exploration, fused GEMM epilogues, row-wise softmax, and FlashAttention. Throughout the tutorial, we work directly with Ti
One fallen power line exposed a growing AI data center problem. Here’s how to fix it.
A close call in Northern Virginia revealed just how poorly data centers respond to grid disruptions. Here's how to fix the problem.
.jpg%3Fwidth%3D720%26quality%3D80%26disable%3Dupscale&s=kAmg5aaORKYzlMWL0upGuDAjzwWxVE-6xoVpfMYWKf4)
