Skip to content

Pull requests: pytorch/FBGEMM

New pull request New
424 Open 3,299 Closed
424 Open 3,299 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

Better kernel launch utilities cla signed fb-exported
#3914 opened Apr 2, 2025 by q10 Loading…
3
Preshuffled BF16I4 Gemm Kernel cla signed fb-exported
#3913 opened Apr 2, 2025 by jwfromm Loading…
5
Add EmbeddingSpMDM8Bit_Sve cla signed fb-exported
#3911 opened Apr 1, 2025 by Nicoshev Loading…
2
[fbgemm_gpu] Enable ROCm builds for GenAI ciflow/rocm cla signed module: rocm
#3910 opened Apr 1, 2025 by q10 Loading…
2
Debug stochastic rounding issue 2 cla signed fb-exported
#3907 opened Apr 1, 2025 by sryap Loading…
2
Debug stochastic rounding issue cla signed fb-exported
#3906 opened Apr 1, 2025 by sryap Loading…
3
Enable slow accumulation in fp8 grouped gemm cla signed fb-exported
#3904 opened Mar 31, 2025 by jwfromm Loading…
2
support permute_multi_embedding_function on torch.export cla signed fb-exported
#3897 opened Mar 28, 2025 by zejunh Loading…
7
Debug A100 too many resources requested for launch issue cla signed fb-exported
#3893 opened Mar 28, 2025 by jianyuh Loading…
2
Add NEON transpose kernel for half-precision cla signed
#3892 opened Mar 28, 2025 by skykongkong8 Loading…
1
[fbgemm_gpu] Update Nova jobs cla signed
#3890 opened Mar 27, 2025 by q10 Loading…
2
[fbgemm_gpu][DO NOT MERGE] Build for vLLM cla signed
#3887 opened Mar 26, 2025 by q10 Loading…
2
Integrate D71065405 and D71079311 into stochastic rounding cla signed fb-exported
#3882 opened Mar 26, 2025 by q10 Loading…
2
nested dispatching of segment_csr on cpu/gpu cla signed fb-exported
#3881 opened Mar 26, 2025 by jeetkanjani7 Loading…
3
Improve Fused8BitRowwiseQuantizedSBFloatToFloatOrHalfNeon by 2%-10% cla signed fb-exported
#3879 opened Mar 25, 2025 by Nicoshev Loading…
2
int4 kv cla signed fb-exported
#3878 opened Mar 25, 2025 by Aya-ZIbra Loading…
2
Back out "Replace LR access with wrapper" cla signed fb-exported
#3857 opened Mar 20, 2025 by spcyppt Loading…
2
Fix CUDA kernel index data type in deeplearning/fbgemm/fbgemm_gpu/fb/src/programmable_kernel/op_merge_bucketized_dense.cuh +10 cla signed fb-exported
#3843 opened Mar 18, 2025 by r-barnes Loading…
3
Refactoring of NoPE cla signed fb-exported
#3840 opened Mar 17, 2025 by Aya-ZIbra Loading…
3
Cleanups for the EEG-based TBE benchmark CLI, pt 3 cla signed fb-exported
#3837 opened Mar 17, 2025 by q10 Loading…
2
Revert D71179541: Multisect successfully blamed "D71179541: [fmoe] update the sorting kernel for bf16 ck fmoe kernel" for one test failure cla signed fb-exported
#3833 opened Mar 17, 2025 by zjing14 Loading…
2
Comment out cache_weights assignment cla signed fb-exported
#3824 opened Mar 15, 2025 by sryap Loading…
2
Set cache_precision = FP16 cla signed fb-exported
#3823 opened Mar 15, 2025 by sryap Loading…
2
Unifying TBE API using List (Frontend) - reland ci-no-td cla signed fb-exported
#3821 opened Mar 14, 2025 by spcyppt Loading…
6
[CUTLASS] Roll cutlass version back a bit to hopefully fix compilation errors. cla signed
#3816 opened Mar 14, 2025 by jwfromm Loading…
2
ProTip! no:milestone will show everything without a milestone.