# PyTorch Release 2.12 | Key Dates

**URL:** <https://dev-discuss.pytorch.org/t/pytorch-release-2-12-key-dates/3329>\
**Category:** Release Announcements\
**Created:** [March 27, 2026, 5:02pm UTC](https://dev-discuss.pytorch.org/t/pytorch-release-2-12-key-dates/3329 "2026-03-27T17:02:54Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![yangw-dev](https://yyz2.discourse-cdn.com/flex036/user_avatar/dev-discuss.pytorch.org/yangw-dev/32/3194_2.png) [@yangw-dev](https://dev-discuss.pytorch.org/u/yangw-dev)\
**Post date:** [March 27, 2026, 5:02pm UTC](https://dev-discuss.pytorch.org/t/pytorch-release-2-12-key-dates/3329/1 "2026-03-27T17:02:54Z")

</div>

Hi Team,

We are kicking off the PyTorch 2.12 release cycle and continue to be excited for all the great features from our PyTorch community!

WHEN ARE THE CRITICAL DATES FOR THE PYTORCH 2.12 RELEASE?

- M1: Release Announcement (27/3/26)

- M2: All PRs landed in PyTorch repo / Feature Submission Closed (10/4/26)

- M3.1: Release branch cut (13/4/26 - 15/4/26)

- M4: Release branch finalized, Announce final launch date, Feature classifications published (week of 27/4/26) - Final RC is produced.

- M4.1: Tutorial drafts submission deadline (6/5/26)

- M5: External-Facing Content Finalized (8/5/26)

- M6: Release Day (13/5/26)

FEATURE TRACKING & CLASSIFICATION

As mentioned in this[RFC](https://github.com/pytorch/pytorch/issues/152134) beginning in release 2.8, feature submissions for the release are tracked as GitHub Issues. You can either create a new issue or [tag an existing RFC](https://github.com/pytorch/pytorch/issues?q=state%3Aopen%20label%3Arelease-feature-request) as to include it in the upcoming release. If you would like a feature to be included in the release blogpost, please mention it in the “New Feature for Release” issue that you create, and include the release version this is targeted to. Please also read the [blog post for the 2.11 release](https://pytorch.org/blog/pytorch-2-11-release-blog) to get familiar with the marketing format.

Also since 2.8 are updates to classification i.e. features are now either Stable (API-Stable) or Unstable (API-Unstable). The previous classifications of Prototype, Beta and Stable, will no longer be used. The requirements for a feature to be considered stable remain the same, and [in the RFC we propose a suggested path to stable](https://github.com/pytorch/pytorch/issues/152134)

If you still have questions after reading the material –or at any time during the process–please don’t hesitate to contact one of us: Gregory Chanan, Alban Desmaison, Andrey Talman, Nikita Shulga, Eli Uriegas, Yang Wang.

Cheers,

Team PyTorch

#oneteam