Skip to content

Pull requests: ml-explore/mlx

New pull request New
43 Open 2,072 Closed
43 Open 2,072 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

Update nanobind version to v2.12.0
#3396 opened Apr 11, 2026 by jrp2014 Loading…
4 tasks
Add clear_streams API for cleanup before exit
#3395 opened Apr 11, 2026 by zcbenz Collaborator Loading…
fix: There is no Stream(gpu, {}) in current thread
#3391 opened Apr 10, 2026 by kyr0 Loading…
1 of 4 tasks
2
Fix JACCL GID index for multi-node Thunderbolt 5 setups
#3389 opened Apr 9, 2026 by qubitcontracting Loading…
4 tasks done
Fix SDPA vmap with GQA/MQA shapes (n_heads != n_kv_heads)
#3385 opened Apr 8, 2026 by Brooooooklyn Contributor Loading…
4 tasks done
1
[CUDA] Handle residue k in qmm_naive
#3379 opened Apr 6, 2026 by zcbenz Collaborator Loading…
Add basic fp8/bfp8 support, future implementations of 8-bit floating …
#3374 opened Apr 4, 2026 by Geramy Loading…
3 tasks done
1
Validate safetensors data offsets
#3364 opened Apr 4, 2026 by MillaFleurs Contributor Loading…
4 tasks done
5
Prevent out-of-bounds memory access caused by corrupt tensor.ndim in gguf file
#3359 opened Apr 3, 2026 by MillaFleurs Contributor Loading…
4 tasks done
14
Add TurboQuant KV cache compression with native Metal SDPA kernel
#3328 opened Mar 28, 2026 by arozanov Loading…
4 tasks done
5
add nn.WeightNorm layer
#3296 opened Mar 22, 2026 by mm65x Contributor • Draft
4 tasks done
[Metal] Fused Flash Attention backward (VJP) kernels
#3241 opened Mar 11, 2026 by Brooooooklyn Contributor Loading…
1
Add bias support to QQLinear
#3215 opened Mar 6, 2026 by mdepree Loading…
4 tasks done
7
Fix shapeless compile eliding reductions on size-1 dimensions
#3202 opened Mar 4, 2026 by ghstrider Loading…
1
2
Add bessel_i0e and bessel_i1e ops
#3193 opened Mar 3, 2026 by robert-johansson Contributor Loading…
2 of 3 tasks
Fix command buffer memory tracking to use bytes instead of elements
#3192 opened Mar 3, 2026 by hxu296 Contributor Loading…
3 of 4 tasks
2
Add lgamma and digamma ops
#3181 opened Feb 27, 2026 by robert-johansson Contributor Loading…
3 of 4 tasks
1
Add all_to_all collective primitive
#3164 opened Feb 24, 2026 by 0xDaizz Loading…
2 of 4 tasks
1
Add 1-bit affine quantization support (Metal)
#3161 opened Feb 24, 2026 by khosravipasha Loading…
4 tasks done
23
Add Expert Parallelism for MoE inference
#3158 opened Feb 23, 2026 by 0xDaizz • Draft
1 of 7 tasks
3
[CUDA] columnwise quantize with tma
#3157 opened Feb 23, 2026 by nastya236 Collaborator Loading…
7
feat: native i0 (modified Bessel function) and kaiser window
#3156 opened Feb 22, 2026 by Vlor999 Contributor Loading…
4 tasks done
[JACCL] Fix silent data corruption from unchecked RDMA work completion status
#3152 opened Feb 21, 2026 by 0xDaizz Loading…
6
[CUDA][Performance] Add radix select implementation for efficient partition operations
#3117 opened Feb 9, 2026 by Lyxot Contributor Loading…
3 of 4 tasks
28
Prevent allocator from operating on residency set during metal capture
#3114 opened Feb 8, 2026 by robertmsale • Draft
3 of 4 tasks
6
ProTip! Type g i on any issue or pull request to go back to the issue listing page.