Bump nanobind from 2.10.2 to 2.13.0

[dependabot]

Bumps nanobind from 2.10.2 to 2.13.0.

Changelog

Sourced from nanobind's changelog.

Version 2.13.0 (Jun 18, 2026)

This release bundles a large set of performance improvements on nanobind's critical paths. It makes object construction 1.42×-3.2× faster, accelerates function dispatch 1.2×-1.5× and accelerates exchange of ndarrays by up to 2.4×. The release also hardens nanobind against error conditions, and fixes race conditions in free-threaded Python builds.

• Performance improvements:

  • Added the :cpp:class:nb::pooled() <pooled> class binding annotation, which maintains a per-type pool of instances to accelerate workloads that create large numbers of short-lived objects. Released objects are stashed in the pool and cheaply recycled, skipping allocation, instance registration, and locking (on free-threaded builds). A microbenchmark exercising object construction runs between 1.42× (regular Python) to 3.2× (free-threading with contention) faster. (PR [#1366](https://github.com/wjakob/nanobind/issues/1366) <https://github.com/wjakob/nanobind/pull/1366>, commit 962cdf <https://github.com/wjakob/nanobind/commit/962cdf735984166ba13bbb5c729c2fe4aaa363be>).

  • A new "medium" function dispatcher accelerates calls to functions whose arguments are merely named or carry default values (and that use neither :cpp:class:nb::args <args>/:cpp:class:nb::kwargs <kwargs> nor more than 8 arguments). Such functions previously fell back to the fully general dispatcher. Positional calls to them now run roughly 32% faster, nearly matching positional-only functions. (PR [#1370](https://github.com/wjakob/nanobind/issues/1370) <https://github.com/wjakob/nanobind/pull/1370>__).

  • A specialized fast path now accelerates simple two-argument calls such as binary operators and copy constructors. (PR [#1362](https://github.com/wjakob/nanobind/issues/1362) <https://github.com/wjakob/nanobind/pull/1362>__).

  • A careful tuning pass on the critical path of function calls and object construction led to speedups ranging from 6.2% on regular builds to 16.4% on stable ABI builds. (PR [#1374](https://github.com/wjakob/nanobind/issues/1374) <https://github.com/wjakob/nanobind/pull/1374>, with a further immortal-type optimization in commit 82f0ce <https://github.com/wjakob/nanobind/commit/82f0ce42905419979a48349883ae945017ab5803>).

  • Optimized the :cpp:class:nb::ndarray <ndarray> import and export critical path. Returning an array becomes up to ~58% faster, and consuming one up to ~21% faster. (PR [#1375](https://github.com/wjakob/nanobind/issues/1375) <https://github.com/wjakob/nanobind/pull/1375>). Separately, improved __dlpack__() keyword parsing speeds numpy.from_dlpack() by a further ~8% (PR [#1373](https://github.com/wjakob/nanobind/issues/1373) <https://github.com/wjakob/nanobind/pull/1373>).

  • Optimized the :cpp:class:nb::ndarray <ndarray> import and export critical path. Returning an array becomes up to ~58% faster, and consuming one up to ~21% faster. (PR [#1375](https://github.com/wjakob/nanobind/issues/1375) <https://github.com/wjakob/nanobind/pull/1375>__). Separately, improved __dlpack__() keyword parsing speeds

... (truncated)

Commits

• e2dc00f v2.13.0 release
• 6a0fa38 changelog updates
• 148755a ndarray: support returning Apple MLX arrays (nb::ndarray\<nb::mlx>)
• 5201b00 ndarray: exception-free argument detection and detection cleanup
• 07b8de2 Reduce redundant Py_TYPE() calls
• 69b9c2f Added a few more type conversions
• e40d69f Fix Clang implicit conversion warnings, test on CI in the future
• 2acdac6 Fast rejection of non-string/capsule objects
• eb2414e ndarray: accept and ignore the __dlpack__ stream argument
• 7707cca Support nb::pooled() for garbage-collected types
• Additional commits viewable in compare view

Note
Automatic rebases have been disabled on this pull request as it has been open for over 30 days.

[PiotrPich2024]

@dependabot rebase

[grzanka]

@dependabot rebase