Skip to content

Pull requests: ml-explore/mlx

New pull request New
47 Open 2,762 Closed
47 Open 2,762 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Fix setitem: negative index normalized against wrong axis size when preceded by None
#4414 opened Aug 27, 2026 by Adityaj0 Contributor Loading…
3 of 4 tasks
Fix a deadlock after test succeed due to mx.clear_streams() holding GIL while synchronously waiting for Metal
#4413 opened Aug 27, 2026 by aleroot Contributor Loading…
5
Pack gated_delta_seq: 8 value rows per SIMD-group
#4409 opened Aug 26, 2026 by wyanzhao Contributor Loading…
Make concurrency cap on load adaptative so that I/O scales with the machine
#4408 opened Aug 26, 2026 by aleroot Contributor Loading…
4
fix(metal): revert inactive NAX MoE simdgroup skipping
#4404 opened Aug 25, 2026 by jonathan308 Contributor • Draft
2
Add support for matrix transpose await verification This pull request is non-trivial and requires a human expert to verify its correctness. low priority
#4402 opened Aug 25, 2026 by aaishwarymishra Contributor Loading…
7
[CUDA] Use native events for GPU fence waits
#4401 opened Aug 25, 2026 by strayberry Contributor Loading…
1
6
Clamp slice bounds in mixed fancy/slice indexing await response This pull request is waiting for response from the author. low priority
#4397 opened Aug 25, 2026 by Adityaj0 Contributor Loading…
1
1
Fix crash on ellipsis indexing with too many trailing indices await response This pull request is waiting for response from the author. low priority
#4396 opened Aug 25, 2026 by Adityaj0 Contributor Loading…
1
4
Fix non transposed affine qmm dispatch logic
#4392 opened Aug 25, 2026 by RohanGautam Contributor Loading…
4
[CUDA] Grouped mm
#4390 opened Aug 24, 2026 by nastya236 Collaborator • Draft
Add explicit index and bytes support low priority
#4388 opened Aug 24, 2026 by aaishwarymishra Contributor Loading…
1
Fix floating point promotion in arithmetic reduction low priority
#4387 opened Aug 23, 2026 by Ved235 Contributor Loading…
2 of 4 tasks
Replace hand-rolled axis normalization with normalize_axis_index in the scan ops await verification This pull request is non-trivial and requires a human expert to verify its correctness. low priority
#4383 opened Aug 23, 2026 by Adityaj0 Contributor Loading…
2
Extending commit for PR#4077 for GQA size 12 and 16 await verification This pull request is non-trivial and requires a human expert to verify its correctness.
#4380 opened Aug 23, 2026 by dudududukim Contributor Loading…
Validate GGUF tensor dimensions low priority
#4378 opened Aug 23, 2026 by roshaninfordham Loading…
6 tasks done
1
Fix mx.from_fp8 E4M3FN NaN decode with branchless carry await verification This pull request is non-trivial and requires a human expert to verify its correctness. bug low priority
#4376 opened Aug 22, 2026 by saud5150 Loading…
1
3
Add fused reduction support await discussion This pull request makes some major changes that requires the team to have a discussion. await verification This pull request is non-trivial and requires a human expert to verify its correctness. enhancement
#4369 opened Aug 21, 2026 by sakupan102 Loading…
1
2
change default axis value to -1 in mlx.core.take_along_axis low priority
#4368 opened Aug 21, 2026 by aaishwarymishra Contributor Loading…
Prevent zero-copy adoption across dtype conversion low priority
#4367 opened Aug 21, 2026 by declanhealy2 Contributor • Draft
Collapse the non sorted axes to pick the contiguous sort kernel await verification This pull request is non-trivial and requires a human expert to verify its correctness. low priority
#4366 opened Aug 21, 2026 by kapellirohith Contributor • Draft
3 of 4 tasks
Fix pad with an axes subset and negative axes low priority
#4364 opened Aug 20, 2026 by kapellirohith Contributor Loading…
3 of 4 tasks
Give put_along_axis axis a default of None like take_along_axis low priority
#4360 opened Aug 19, 2026 by ayaangazali Contributor Loading…
Add nanobind 3.0 split-mode support await discussion This pull request makes some major changes that requires the team to have a discussion.
#4359 opened Aug 19, 2026 by XXXXRT666 Contributor Loading…
4 tasks done
2
Fix integer pow zeroing a whole SIMD vector on a negative exponent low priority
#4354 opened Aug 19, 2026 by ayaangazali Contributor Loading…
ProTip! Updated in the last three days: updated:>2026-08-24.