Skip to content

Pull requests: pytorch/FBGEMM

New pull request New
395 Open 3,114 Closed
395 Open 3,114 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
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Add cublas FP8 tensorwise GEMM in fbgemm quantize bench cla signed fb-exported
#3693 opened Feb 14, 2025 by jiawenliu64 Loading…
2
[fbgemm_gpu] Test genai op registration cla signed
#3692 opened Feb 14, 2025 by q10 Loading…
1
avoid extra copy in PackedGemmMatrixB constructor cla signed fb-exported
#3691 opened Feb 13, 2025 by helloguo Loading…
4
[fbgemm_gpu] Increase timeout for ARM nova jobs cla signed
#3690 opened Feb 13, 2025 by q10 Loading…
1
Numerical Fix. cla signed fb-exported
#3688 opened Feb 13, 2025 by levendlee Loading…
2
custom reduce scatter cla signed fb-exported
#3686 opened Feb 13, 2025 by xw285cornell Loading…
3
adding an option to skip zeroing output tensor for f8f8bf16_rowwise_grouped_dynamic cla signed fb-exported
#3685 opened Feb 13, 2025 by mxz297 Loading…
2
Add small M support cla signed fb-exported
#3682 opened Feb 12, 2025 by YUNQIUGUO Loading…
3
[wip] commits scraper cla signed
#3676 opened Feb 11, 2025 by q10 Loading…
1
Add D_folded support for jagged_to_padded_dense_backward meta function cla signed fb-exported
#3670 opened Feb 8, 2025 by brad-mengchi Loading…
2
changing config for fp8 gemm cla signed fb-exported
#3668 opened Feb 7, 2025 by adamomainz Loading…
3
fixing fp8 gemm on amd cla signed fb-exported
#3662 opened Feb 5, 2025 by adamomainz Loading…
2
Pull ARM's matrix transpose PR cla signed fb-exported
#3660 opened Feb 4, 2025 by Nicoshev Loading…
2
Adding Missing includes and explicitly declaring Tensor in aten namespace. cla signed fb-exported
#3638 opened Jan 30, 2025 by pradeepfn Loading…
2
Partial revert of D66986498 (Optimized backward pass for ROCm devices, pt 1), 2nd attempt ciflow/rocm cla signed fb-exported module: rocm
#3637 opened Jan 29, 2025 by q10 Loading…
3
avoid using warning tensor in cpu tbe op cla signed fb-exported
#3631 opened Jan 29, 2025 by 842974287 Loading…
2
Update bf16i4 gemm with new cutlass version cla signed fb-exported
#3630 opened Jan 29, 2025 by jwfromm Loading…
2
finish #1808 cherry-pick, adjust interface cla signed fb-exported
#3627 opened Jan 28, 2025 by coconutruben Loading…
2
Partial revert of D66986498 cla signed fb-exported
#3620 opened Jan 27, 2025 by q10 Loading…
2
Re-land D67407935 (Optimized backward pass for ROCm devices, pt 2) ciflow/rocm cla signed fb-exported module: rocm
#3619 opened Jan 27, 2025 by q10 Loading…
3
Performance Optimization: Optimized TileShape Configuration for f8 cla signed
#3617 opened Jan 27, 2025 by MatrixAssembler Loading…
7
Replace runners prefix amz2023. (#2895) cla signed fb-exported module: rocm
#3612 opened Jan 24, 2025 by q10 Loading…
2
Test out D68193920 cla signed fb-exported
#3606 opened Jan 23, 2025 by q10 Loading…
7
AdagradW cla signed fb-exported
#3605 opened Jan 23, 2025 by spcyppt Loading…
11
[fbgemm_gpu] Try nested ops namespaces cla signed
#3603 opened Jan 22, 2025 by q10 Loading…
1
ProTip! no:milestone will show everything without a milestone.