Google has joined Amazon in attacking Nvidia’s stronghold with AI app developers

AI development tools now offer direct pathways to run on Google and Amazon AI chips, not just Nvidia's.

Google has open-sourced TorchTPU, a version of the popular PyTorch AI development toolkit designed to run on Google's Tensor Processing Units (TPUs). This move, announced October 20, 2026, allows AI applications to leverage Google's custom AI chips and Google Cloud with minimal configuration changes, without requiring developers to rewrite their code. TorchTPU's availability follows a similar initiative from Amazon, which launched TorchNeuron to enable PyTorch applications to run on Amazon's Trainium AI chips and AWS. Both companies are challenging Nvidia's dominant position in the AI hardware market, as most AI applications are currently optimized for Nvidia's GPUs. Google claims enhanced compatibility with other AI tools through collaboration with Meta and other industry players, aiming to facilitate easier migration for developers.

Google's open-sourcing of TorchTPU, in collaboration with key AI development platforms, signifies a strategic effort to dismantle Nvidia's de facto hardware monopoly for AI applications. By providing a PyTorch variant optimized for its TPUs and ensuring broad compatibility with popular AI libraries, Google aims to make its cloud infrastructure a more attractive and accessible choice for AI developers. This move directly challenges Nvidia's dominance, which has been built on the widespread adoption of PyTorch with GPU-specific optimizations.

Amazon's parallel launch of TorchNeuron and its contribution to decoupling PyTorch from Nvidia-specific features further underscore a concerted push by major cloud providers to offer viable alternatives to Nvidia's hardware. Developers may soon find it increasingly seamless to deploy AI models on cloud-agnostic or provider-specific AI silicon, potentially leading to more competitive pricing and specialized hardware offerings. The ease of porting existing PyTorch applications will be a key factor in adoption.