According to reports, Google is researching to make TPU more compatible with the AI software framework PyTorch in an effort to become a viable alternative to NVIDIA GPUs

AASTOCKS
2025.12.18 00:58

According to a report by Reuters, Google, a subsidiary of Alphabet (GOOGL.US), is promoting a project codenamed "TorchTPU" to make its Tensor Processing Unit (TPU) artificial intelligence (AI) chips better compatible with PyTorch, the most widely used AI software framework globally, and fully compatible with customers who have built their technology architecture based on PyTorch, becoming a viable alternative to NVIDIA (NVDA.US) graphics processing units (GPUs). The report indicates that Google is also considering open-sourcing some software to accelerate customer adoption.

PyTorch is an open-source project supported by Meta Platforms (META.US) and is one of the most commonly used tools by engineers developing AI models. The project is a set of pre-written code libraries and frameworks that can automate many common tasks in AI software development. Since its release in 2016, PyTorch's development has been closely tied to NVIDIA's CUDA software, which the market views as NVIDIA's strongest moat against competitors

Source: AASTOCKS The copyright of this article belongs to the original author/organization.
The views expressed herein are solely those of the author and do not reflect the stance of the platform. The content is intended for investment reference purposes only and shall not be considered as investment advice. Please contact us if you have any questions or suggestions regarding the content services provided by the platform.
© 2026 Longbridge｜Disclaimer