Skip to content

Pull requests: pytorch/FBGEMM

New pull request New
519 Open 3,985 Closed
519 Open 3,985 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

Replace .packed_accessor32() with PTA_B() in intraining embedding pruning ops cla signed fb-exported
#4718 opened Aug 16, 2025 by q10 Loading…
2
fix typo: quantize_packed_fp8_symmetric -> dequantize_packed_fp8_symmetric (#1740) cla signed fb-exported
#4717 opened Aug 15, 2025 by ColinPeppler Loading…
3
Extract the res backend to a separate class and export to python side cla signed fb-exported
#4714 opened Aug 15, 2025 by chouxi Loading…
4
Add unit tests for Full Adam, pt 2 cla signed fb-exported
#4713 opened Aug 15, 2025 by q10 Loading…
2
Enable MXFP8 grouped GEMM cla signed fb-exported
#4710 opened Aug 15, 2025 by jiawenliu64 Loading…
3
Fix quantize kernels on rocm 6.4 ciflow/rocm cla signed fb-exported module: rocm
#4708 opened Aug 15, 2025 by jwfromm Loading…
5
Add logging in kv_dram init cla signed fb-exported
#4706 opened Aug 14, 2025 by chenyuzhcy Loading…
2
[ROCm] remove hipify work-around that is no longer needed cla signed module: rocm
#4705 opened Aug 14, 2025 by jeffdaily Loading…
2
Enable USE_FBGEMM_GENAI cla signed fb-exported
#4703 opened Aug 14, 2025 by cthi Loading…
2
Fix fp32 ARM builds cla signed
#4699 opened Aug 13, 2025 by cyyever Loading…
9
Support optimizer state offloading for the full Adam optimizer cla signed fb-exported
#4696 opened Aug 13, 2025 by q10 Loading…
2
Boost performance of MXFP4 quantization with inline PTX cla signed fb-exported
#4694 opened Aug 13, 2025 by jiawenliu64 Loading…
2
emitTB metrics + bug fix cla signed fb-exported
#4687 opened Aug 12, 2025 by kathyxuyy Loading…
2
Fold unaligned vec4 load and store into function cla signed fb-exported
#4684 opened Aug 12, 2025 by q10 Loading…
2
Feature score eviction frontend support cla signed fb-exported
#4682 opened Aug 12, 2025 by EddyLXJ Loading…
6
Adding trace to inference benchmark cla signed fb-exported
#4674 opened Aug 11, 2025 by gchalump Loading…
4
Add TBE data configuration reporter to TBE forward (v3) (#4455) cla signed fb-exported
#4672 opened Aug 11, 2025 by gchalump Loading…
6
Allowing small tiles to work on 2k*2k shapes cla signed fb-exported
#4669 opened Aug 11, 2025 by RandySheriff Loading…
2
Dummy commit to test for errors cla signed fb-exported
#4668 opened Aug 10, 2025 by ionuthristodorescu Loading…
2
Migrate sparse ops kernels to FBGEMM_LAUNCH_KERNEL, pt 7 cla signed fb-exported
#4653 opened Aug 7, 2025 by q10 Loading…
2
implement jagged_unique_indices_cpu cla signed fb-exported
#4651 opened Aug 7, 2025 by wangyh Loading…
2
Adding performance logger calls cla signed fb-exported
#4650 opened Aug 7, 2025 by ionuthristodorescu Loading…
13
Add 'device-with-speclist' bench cla signed fb-exported
#4648 opened Aug 6, 2025 by YanXiong-Meta Loading…
9
Add exhaustive autotune results for 500x IGCTR cla signed fb-exported
#4644 opened Aug 6, 2025 by JChunX Loading…
6
Adding optional D multiplication to index shuffling cla signed fb-exported
#4631 opened Aug 1, 2025 by sunfish2010 Loading…
2
ProTip! Follow long discussions with comments:>50.