●Change8

Migrating to PyTorch v2.13.0

Version v2.13.0 introduces 8 breaking changes. This guide details how to update your code.

Released: 7/8/2026

8
Breaking Changes
8
Migration Steps
15
Affected Symbols

⚠️ Check Your Code

If you use any of these symbols, you need to read this guide:

PyObjecttorch._C.parse_schemaStorageImplStorageImpl::is_cow()StorageImpl::maybe_materialize_cow()cow::materialize_cow_storage()StorageImpl::set_materializer()StorageImpl::has_materializer()StorageImpl::clear_materializer()Tensor::grad_fnstd::shared_ptr<Node>c10::intrusive_ptr<Node>Tensor.namesTensor.rename()Tensor.refine_names

Breaking Changes

●Issue #1

Stop building CPython 3.13t (free-threaded) binaries. Users on the free-threaded interpreter should move to Python 3.14t.

●Issue #2

Bare `PyObject` is no longer allowed in operator schemas. Parsing a schema with a bare `PyObject` argument or return type will now raise a schema parse error.

●Issue #3

Remove Bazel build support. Users building PyTorch with Bazel should migrate to the supported CMake/pip install flow.

●Issue #4

Enforce C++20 minimum in header guards. C++20 is now required to import PyTorch headers.

●Issue #5

`StorageImpl`'s built-in copy-on-write (COW) materialization is replaced by a pluggable materializer hook. Out-of-tree backends/extensions calling removed COW symbols (`is_cow()`, `maybe_materialize_cow()`, `cow::materialize_cow_storage()`) will fail to compile. Migrate to the new hook API (`set_materializer()` / `has_materializer()` / `clear_materializer()`).

●Issue #6

Convert `shared_ptr<Node>` to `intrusive_ptr<Node>` in autograd. The signature of `Tensor.grad_fn()` changes from `std::shared_ptr<Node>` to `c10::intrusive_ptr<Node>`. Construction of C++ autograd functions must use `c10::make_intrusive` instead of `std::shared_ptr` with a custom deleter.

●Issue #7

The minimum supported NCCL version when building from source is now 2.23. Users building from source against an older NCCL will hit compile errors.

●Issue #8

Remove named tensors. All associated Python and C++ APIs, including `Tensor.names`, `Tensor.rename()`, and `Tensor.refine_names`, are removed.

Migration Steps

  1. 1
    If using free-threaded Python 3.13t, move to Python 3.14t.
  2. 2
    If parsing operator schemas, ensure bare `PyObject` is not used as an argument or return type; use specific types or handle schema parsing errors.
  3. 3
    If building from source, migrate from Bazel build system to the CMake/pip install flow.
  4. 4
    If compiling C++ extensions, ensure your compiler is set to C++20 standard or newer.
  5. 5
    If out-of-tree backends/extensions directly called COW entry points on `StorageImpl` (e.g., `is_cow()` or `maybe_materialize_cow()`), migrate to using the pluggable materializer hook API (`set_materializer()` / `has_materializer()` / `clear_materializer()`).
  6. 6
    In C++ code accessing `Tensor.grad_fn()`, change usage from `std::shared_ptr<Node>` to `c10::intrusive_ptr<Node>`. When constructing C++ autograd nodes, use `c10::make_intrusive<CustomCppNode>()` instead of `std::shared_ptr` with a custom deleter.
  7. 7
    If building from source, ensure the system NCCL installation is version 2.23 or newer.
  8. 8
    If using named tensors APIs (e.g., `Tensor.names`, `Tensor.rename()`), update code as these features have been removed.

Release Summary

PyTorch 2.13 introduces significant performance features like FlexAttention on MPS and the CuTeDSL backend for Inductor, alongside major internal cleanups including the removal of Bazel support and named tensors. This release contains several breaking changes, particularly around C++ APIs and Python interpreter version support.

Need More Details?

View the full release notes and all changes for PyTorch v2.13.0.

View Full Changelog