Skip to content

Pull requests: ml-explore/mlx

New pull request New
47 Open 2,105 Closed
47 Open 2,105 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

feat(metal): add fused Mamba-3 SSD kernel for sub-quadratic sequence modeling
#3517 opened May 11, 2026 by diysumit Loading…
mlx launch clean
#3513 opened May 10, 2026 by nastya236 Collaborator Loading…
Add copy keyword to mx.asarray
#3510 opened May 10, 2026 by eyupcanakman Loading…
4 tasks done
1
[CUDA] Guard qmm_naive scale and bias loads at tile boundaries
#3509 opened May 10, 2026 by Lyxot Contributor Loading…
1
2
[CUDA] Fix gather_mm
#3503 opened May 10, 2026 by zcbenz Collaborator Loading…
[CPU - Linux] AVX SIMD backend for fp16 and bf16 matmul
#3502 opened May 9, 2026 by acsweet Loading…
3 of 4 tasks
1
Pass deployment target when linking metallib
#3501 opened May 9, 2026 by dhiltgen Contributor Loading…
2 of 4 tasks
1
feat(utils): add register_pytree_node and pytree-aware compile wrapper
#3500 opened May 8, 2026 by st-adam Loading…
4 tasks done
1
1
Fix rope single token multiple sequences
#3498 opened May 8, 2026 by angeloskath Member Loading…
1
1
Fix qvm_split_k incorrect batch stride calculation
#3497 opened May 8, 2026 by angeloskath Member Loading…
1
1
Improve DLPack-compatible array imports
#3495 opened May 8, 2026 by XXXXRT666 Contributor Loading…
4 tasks done
8
Fix scatter_prod GPU hang on NaN with contention
#3492 opened May 7, 2026 by tillahoffmann Loading…
3 tasks done
7
Use uv in macOS CI
#3491 opened May 7, 2026 by zcbenz Collaborator Loading…
test: Accumulate CPU half-precision sums in float32
#3488 opened May 6, 2026 by sofinvalery Contributor Loading…
4 tasks done
1
9
feat(compile): CustomKernel and GatherQMM implement output_shapes for shapeless compile
#3485 opened May 5, 2026 by dexwritescode Loading…
10
[Python] Fix mx.array DLPack dispatch
#3476 opened May 4, 2026 by apstenku123 Loading…
9
Fix JACCL GID selection on Apple Thunderbolt RDMA (errno 22 RTR fail)
#3468 opened Apr 29, 2026 by danielkristofik Loading…
3 of 4 tasks
1
Add MLX_SDPA_BLOCKS env var for 2-pass vector kernel block-count override
#3455 opened Apr 26, 2026 by adurham Loading…
7
[jaccl] Fix race on local_staging in MeshImpl::all_reduce
#3451 opened Apr 25, 2026 by kernelpool Contributor Loading…
3 of 4 tasks
1
Handle non-multiple-of-8 spatial dims in depthwise conv2d Metal path
#3446 opened Apr 24, 2026 by qflen Contributor Loading…
4 tasks done
1
1
Reduce peak memory of large half-precision random.uniform/normal
#3439 opened Apr 22, 2026 by dogukanveziroglu Contributor • Draft
4 tasks done
5
fix: avoid clearing in-flight pipeline states in custom kernel cache
#3434 opened Apr 21, 2026 by Ziqiao-git Contributor Loading…
2
fsdp: improve state saving determinism and tree reconstruction
#3407 opened Apr 13, 2026 by coder-jayp Loading…
1 of 4 tasks
3
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
Add basic fp8/bfp8 support, future implementations of 8-bit floating …
#3374 opened Apr 4, 2026 by Geramy Loading…
3 tasks done
2
ProTip! Add no:assignee to see everything that’s not assigned.