0.22.0 release

Open

No due date

•Last updated Nov 2, 2026

56% complete

Set minimum supported NumPy version to 2.0.0
Unread
#2615 opened Oct 8, 2025 · Updated Oct 12, 2026Draft
2
Integrate dpnp.tensor docs to dpnp
Unread
#2946 opened Jun 11, 2026 · Updated Oct 12, 2026
2
Speed up in-place vector-to-C-contiguous-matrix broadcast on CPU
Unread
#2981 opened Jun 30, 2026 · Updated Oct 12, 2026
1
Add cho_factor and cho_solve to dpnp.scipy.linalg
Unread
#3116 opened Oct 20, 2026 · Updated Oct 30, 2026
9
Refactor libtensor and backend into libtensor
Unread
#3131 opened Oct 29, 2026 · Updated Nov 2, 2026
15
Add dpnp.top_k
Unread
#3133 opened Oct 30, 2026 · Updated Nov 2, 2026
15
Sort csr_matrix indices once on a combined row and column key
Unread
#3134 opened Oct 30, 2026 · Updated Oct 30, 2026
4
Add dpnp.minmax with a fused min/max reduction kernel
Unread
#3135 opened Oct 30, 2026 · Updated Oct 30, 2026
6
Support pickle and copy.deepcopy of dpnp.ndarray
Unread
#3142 opened Nov 1, 2026 · Updated Nov 2, 2026Draft
7
Use blobless clones for full-history checkouts in CI
Unread
#3143 opened Nov 2, 2026 · Updated Nov 2, 2026
6

Navigation Menu