From 5b496ed6926699444ddd2224210c6e4b05b99ef0 Mon Sep 17 00:00:00 2001 From: cloudchrischan Date: Mon, 5 Jan 2026 16:01:56 -0800 Subject: [PATCH] Update README.md --- README.md | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/README.md b/README.md index 930968bc..91fbbcd4 100644 --- a/README.md +++ b/README.md @@ -7,6 +7,12 @@

+> [!NOTE] +> `torchprime` is no longer under active development. We are working on a +> [new framework](https://github.com/pytorch/xla/issues/9684) for +> PyTorch on TPU. See [MaxText](https://github.com/AI-Hypercomputer/maxtext) +> for pre-training and post-training with JAX on TPU. + `torchprime` is a reference implementation for training PyTorch models on TPU. It is designed to showcase best practices for large-scale, high-performance model training using `torch_xla` ([project][torch_xla]), with