Skip to content

Enable ckpt deserialization using colocated python. - #1362

Closed
raytomato wants to merge 5 commits into
apple:mainfrom
raytomato:cliu74/colocated_python
Closed

Enable ckpt deserialization using colocated python.#1362
raytomato wants to merge 5 commits into
apple:mainfrom
raytomato:cliu74/colocated_python

Conversation

@raytomato

Copy link
Copy Markdown
Contributor
  1. Cherry picked PR 2/n to 4/n.

* [2/n] Add colocated python path to checkpoint manager.

* Fix threading and asyncio bugs.

* resolve dex-ai's comments.

* Minor, remove redundant logging.
* [3/n] Use limiter to control cpu to tpu transfer to avoid OOM.

* Use explicit profile call to trigger the patch properly.

* Add tensor statics in the benchmark script.

* Add more logging for located python.

* Add accurate compute for per device bytes.
…ther reduce memory usage (#2214)

* [4/n] Use global limiter to control the full model loading pipeline to further reduce memory usage.

* Fix the index array issue. Plus minor optimization of the thread pool.

* Fix parallel execution of colocated python.

* more accurate byte per device computation.

* Enable colocated python by default when it's available. Controlled through ENV when it's available. Controlled through ENV..

* simplify limiter usage in load_and_transfer_one

* Resolve devx-ai's comments.

* fix global_shape issue

* resolve Ethan's comment.

* minor tensorspec open improvement

* resolve Ethan's new comments. And minor improvement in case ts has no read chunk.
@raytomato
raytomato requested review from a team as code owners March 3, 2026 23:35

@LabTopProductions LabTopProductions left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

©️

@github-actions

Copy link
Copy Markdown

This pull request has been automatically marked as stale because it has been inactive for 60 days. It will be closed in 7 days if no further activity occurs. If you would like to continue working on this, please remove the stale label or leave a comment.

@github-actions github-actions Bot added the stale label Jun 28, 2026
@github-actions

github-actions Bot commented Jul 6, 2026

Copy link
Copy Markdown

This pull request was closed because it has been inactive for more than 7 days since being marked as stale. Please feel free to reopen it if you would like to continue.

@github-actions github-actions Bot closed this Jul 6, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants