Skip to content

Pull requests: pytorch/FBGEMM

New pull request New
414 Open 3,200 Closed
414 Open 3,200 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

Backout Unifying TBE API using List (Frontend) cla signed fb-exported
#3803 opened Mar 11, 2025 by spcyppt Loading…
4
Migrate TBE benchmark utilities over to TBE, pt 5 cla signed fb-exported
#3801 opened Mar 11, 2025 by q10 Loading…
5
Add Preshuffled FP8 x INT4 Grouped Gemm Kernel cla signed fb-exported
#3800 opened Mar 11, 2025 by jwfromm Loading…
2
Fix fbgemm_gpu hashtable bench on CPU
#3799 opened Mar 11, 2025 by pnfox Loading…
2
Fix array-bounds error cla signed
#3798 opened Mar 11, 2025 by cyyever Loading…
1
[fbgemm_gpu] Limit the number of ROCm hardware targets ciflow/rocm cla signed module: rocm
#3797 opened Mar 11, 2025 by q10 Loading…
12
Fix jit script cla signed fb-exported
#3796 opened Mar 11, 2025 by spcyppt Loading…
3
Fast BF16 Reduction for AllReduce cla signed fb-exported
#3793 opened Mar 10, 2025 by zjing14 Loading…
7
Log feature gate statuses in TBE init cla signed fb-exported
#3792 opened Mar 10, 2025 by q10 Loading…
4
Performance Optimization: Improved TileShape Configuration for Large Llama Shapes cla signed
#3790 opened Mar 10, 2025 by MatrixAssembler Loading…
1
Enable FP8 Triton dequantized block-wise kernel cla signed fb-exported
#3788 opened Mar 10, 2025 by jiawenliu64 Loading…
3
Fix TBE inference per-sample weight cla signed fb-exported
#3787 opened Mar 10, 2025 by sryap Loading…
3
Handle zero inputs in F8I4 GEMM cla signed fb-exported
#3777 opened Mar 6, 2025 by jiawenliu64 Loading…
2
Add test for VBE CPU cla signed
#3776 opened Mar 6, 2025 by spcyppt Loading…
1
Provide helper functions for int4 quantization cla signed fb-exported
#3775 opened Mar 6, 2025 by jwfromm Loading…
5
Enable KleidiAI for FP16 cla signed fb-exported
#3770 opened Mar 5, 2025 by Nicoshev Loading…
2
[fbgemm_gpu] [WIP] Separate cppstd cla signed
#3768 opened Mar 5, 2025 by q10 Loading…
1
FBGEMM Add Columnwise Weight Scaling to F8I4 GEMM cla signed fb-exported
#3766 opened Mar 5, 2025 by jwfromm Loading…
7
fix: topology_utils.h compilation issue cla signed
#3761 opened Mar 4, 2025 by DevashishLal-CB Loading…
1
Move float conversion functions from Types.h into new FloatConversion.h cla signed fb-exported
#3760 opened Mar 4, 2025 by MatzeB Loading…
3
Invoke CPU schema op for unweighted version TBE on MTIA cla signed fb-exported
#3757 opened Mar 3, 2025 by egienvalue Loading…
3
Enable multi-processing in CPU TBE micro-benchmarks cla signed fb-exported
#3753 opened Feb 28, 2025 by excelle08 Loading…
3
Tuned fp8/bf16 GroupGEMM for MOE_17B cla signed fb-exported
#3752 opened Feb 28, 2025 by zjing14 Loading…
6
Use GEMM kernel for KleidiAI to accelerate FP32Benchmark cla signed
#3751 opened Feb 28, 2025 by milpuz01 Loading…
1
Back out "Move execute_backward_adagrad into a class" cla signed fb-exported
#3749 opened Feb 28, 2025 by q10 Loading…
2
ProTip! Adding no:label will show everything without a label.