Merge pull request #199 from coreweave/eta/long-tensors

feat(serialization): Support tensors with very long dimensions

Author: wbrown

CHANGELOG.md

@@ -5,6 +5,18 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [Unreleased]
+
+### Added
+
+- Tensors with long dimensions (≥ `2 ** 32` elements in a single dimension)
+  can now be serialized and deserialized
+
+### Fixed
+
+- `tensorizer.torch_compat` can now serialize and deserialize tensors that have
+  storages with sizes ≥ `2 ** 32`
+
 ## [2.11.1] - 2025-08-05
 
 ### Fixed
@@ -492,6 +504,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   - `get_gpu_name`
   - `no_init_or_tensor`
 
+[Unreleased]: https://github.com/coreweave/tensorizer/compare/v2.11.1...HEAD
 [2.11.1]: https://github.com/coreweave/tensorizer/compare/v2.11.0...v2.11.1
 [2.11.0]: https://github.com/coreweave/tensorizer/compare/v2.10.1...v2.11.0
 [2.10.1]: https://github.com/coreweave/tensorizer/compare/v2.10.0...v2.10.1

tensorizer/_version.py

@@ -1 +1 @@
-__version__ = "2.11.1"
+__version__ = "2.12.0a0"