Amazon has released TorchNeuron, a developer tool designed to make it easier for creators to run PyTorch—the dominant open-source framework for building AI applications—on Amazon's own chips rather than Nvidia's . The move targets Nvidia's near-monopoly on AI development, where most engineers build and train models using Nvidia hardware.
PyTorch has become the de facto standard for AI development, giving Nvidia outsized influence over which chips developers choose. By creating TorchNeuron, Amazon is attempting to reduce developer lock-in and drive adoption of its custom silicon as a more cost-effective alternative for running AI workloads.
The strategy reflects intensifying competition among cloud providers and chipmakers to capture the fast-growing AI infrastructure market, with Amazon, Google, and others investing heavily in custom chips to reduce reliance on Nvidia.
This is an AI-written summary. The original reporting belongs to TechCrunch Startups. How we write these · Spot a mistake?