Skip to content

Pull requests: ml-explore/mlx

New pull request New
43 Open 2,047 Closed
43 Open 2,047 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

Use metal as the front-end for the metal linker
#3354 opened Apr 1, 2026 by louen Loading…
2 tasks done
Add printoptions
#3333 opened Mar 29, 2026 by ChristophePRAT Loading…
4 tasks done
1
13
fix: fail build when Metal compiler header resolution fails
#3332 opened Mar 29, 2026 by dogukanveziroglu Loading…
2 of 4 tasks
1
Add TurboQuant KV cache compression with native Metal SDPA kernel
#3328 opened Mar 28, 2026 by arozanov Loading…
4 tasks done
5
[CUDA] Add GatherQMM for quantized gather matmul
#3321 opened Mar 25, 2026 by Lyxot Loading…
12
fix: defer error propagation from Metal completion handlers
#3318 opened Mar 25, 2026 by Thump604 Loading…
1
7
Chunked full-attention SDPA for long key sequences
#3307 opened Mar 24, 2026 by Thump604 Loading…
10 tasks done
1
4
Add logsumexp output to fused SDPA kernel
#3306 opened Mar 24, 2026 by Thump604 Loading…
8 tasks done
4
add nn.WeightNorm layer
#3296 opened Mar 22, 2026 by mm65x • Draft
4 tasks done
fix: add head_dim=256 to fused SDPA full attention kernel
#3293 opened Mar 22, 2026 by Thump604 Loading…
23
add causal_upper_left mask option to scaled_dot_product_attention
#3254 opened Mar 14, 2026 by mm65x Loading…
4 tasks done
1
[Metal] Fused Flash Attention backward (VJP) kernels
#3241 opened Mar 11, 2026 by Brooooooklyn Loading…
1
Add bias support to QQLinear
#3215 opened Mar 6, 2026 by mdepree Loading…
4 tasks done
6
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 Loading…
2 of 3 tasks
Fix command buffer memory tracking to use bytes instead of elements
#3192 opened Mar 3, 2026 by hxu296 Loading…
3 of 4 tasks
2
Add lgamma and digamma ops
#3181 opened Feb 27, 2026 by robert-johansson 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
17
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 Loading…
7
feat: native i0 (modified Bessel function) and kaiser window
#3156 opened Feb 22, 2026 by Vlor999 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 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! Adding no:label will show everything without a label.