diff --git a/docs/texts/introduction.md b/docs/texts/introduction.md
index 439a6b2..b694fa6 100644
--- a/docs/texts/introduction.md
+++ b/docs/texts/introduction.md
@@ -47,7 +47,7 @@ Furthermore, **veScale** is also developing a _Mixed Mode_* of partial _Eager_ a
 **veScale** is designed and implemented on top of a primitive called _DTensor_ that provides a global tensor semantic with local shards distributed on multiple devices.
 **veScale** extends and enhances the _PyTorch DTensor_ for our production standard, and further develops the _Auto-Plan*_ and _Auto-Paralleize_ with a unified configuration and API. 
 
-Furthermore, **veScale** also supports online _Auto-Reshard_ for distributed checkpoints, which will be open-sourced as a new project -- **OmniStore**.
+Furthermore, **veScale** also supports online _Auto-Reshard_ for distributed checkpoints.
 
 (`*` is under development)
 
diff --git a/vescale/checkpoint/README.md b/vescale/checkpoint/README.md
index 26b8d1f..b9c0bbe 100644
--- a/vescale/checkpoint/README.md
+++ b/vescale/checkpoint/README.md
@@ -21,8 +21,6 @@ abstracting away the complexities of underlying details such as process rank and
 
 `vescale.checkpoint` incorporates [fast checkpointing](https://arxiv.org/abs/2402.15627) and various I/O optimization techinques, enhancing I/O efficiency during LLM training.  
 
-`vescale.checkpoint` will be a part of `OmniStore` project, a new open-source project coming soon.
-
 `vescale.checkpoint` is built on top of `PyTorch Distributed Checkpoint` with significant differences as discussed above.
 
 ## How to use `vescale.checkpoint`?