Reuters reports roughly 600,000 of Nvidia's H200 accelerators are sitting in Hong Kong while Beijing declines to approve…
Fractile closed roughly $700M with the UK's sovereign AI fund on the cap table while its in-memory inference…
The South Korean inference chip startup reached a $2.9B valuation on the argument that data centers are short…
For the first time, customers can rent Google's tensor processing units outside Google's own data centers.
A rewrite of IBM's compiler moved Spyre onto PyTorch's Inductor backend, cutting the codebase from about 273,000 lines…
TorchTPU swaps a brittle bridge for a dispatcher hook, letting PyTorch models run on TPUs with one device…
Jim Keller's chip company raised $1.5B after showing its Galaxy servers could serve DeepSeek models.
TorchNeuron moves PyTorch workloads onto Trainium with a single settings change, and Amazon is aiming it squarely at…
Nvidia's CEO spent a week in Europe buying stakes in French startups and funding agent safety research.
Talks could end in a purchase, a bigger stake or an acquihire, days after Reflection shipped its first…
Microsoft unveiled Maia 300 and said its homegrown accelerators will move to multi-gigawatt deployments beginning in 2027.
Volantis closed an $88M Series A to commercialize a photonic inference system it says sidesteps the memory wall…