Google Open-Sources TorchTPU and Ships Llm-D 1.0 for Kubernetes AI Inference
Google said TorchTPU lets developers move PyTorch apps to its chips with a simple configuration change and no rewriting.
- On Tuesday, Google launched TorchTPU as an open-source project, enabling developers to run PyTorch applications on Tensor Processing Units with a simple configuration change instead of rewriting code.
- PyTorch, used by 85% of AI apps, was originally developed by Meta in 2016 and transferred to the Linux Foundation, with features optimized for Nvidia GPUs that dominated AI research.
- Google collaborated with Hugging Face, Inferact, and RadixArk to ensure feature compatibility, while launching a dedicated TPU agent suite to help developers migrate applications to Google Cloud.
- Amazon released its own PyTorch version, TorchNeuron, on Monday, allowing AI applications to be ported to its proprietary Trainium chips and AWS without requiring code modifications.
- Both initiatives represent a broader effort by tech giants to break the AI industry's dependence on Nvidia hardware by providing alternatives for running PyTorch models on their own infrastructure.
11 Articles
11 Articles
Google Launches LiteRT Tensor API for Native On-Device ML Development
Google's new LiteRT Tensor API lets developers hand-build on-device models op by op in C++ and run the same graph on CPU, GPU or NPU, or export a .tflite file. Google still recommends converting trained models for most projects; the Tensor API targets developers who want low-level control.
Google, Amazon Target Nvidia’s AI Dominance With New PyTorch Tools for Alternative Chips
Google and Amazon are stepping up efforts to weaken Nvidia’s dominance of the artificial intelligence industry by making it easier for developers to run AI applications on their own chips without rewriting existing software. Google announced on Tuesday that TorchTPU, its open-source implementation of PyTorch for the company’s Tensor Processing Units (TPUs), is now available. […]
Google open-sourced TorchTPU, so PyTorch on TPUs just got a lot easier
Google released TorchTPU as open source, giving PyTorch developers a native way to run models on the company’s Tensor Processing Units. Google also announced TPU support in the vLLM and SGLang serving engines, the TorchTitan training framework, Ray, and Hugging Face’s libraries. Teams that write their models in PyTorch and run them on GPUs by default have the most to gain. Google says those workloads now move to TPUs, its custom AI chips, with m…
Google unveils TorchTPU in a push to break Nvidia's stranglehold on AI compute
Google LLC is striving to pry open Nvidia Corp.’s grip on the artificial intelligence computing industry with the launch of a specialized version of the open-source PyTorch framework, called TorchTPU, which is designed to eliminate a key advantage of its rival. One of the reasons why Nvidia and its graphics processing units remain so dominant […]
Coverage Details
Bias Distribution
- 100% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
