-
Notifications
You must be signed in to change notification settings - Fork 736
Pull requests: pytorch/FBGEMM
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Harden repeat_arange benchmark with input validation and trace export (#5676)
cla signed
fb-exported
meta-exported
#5676
opened Apr 22, 2026 by
q10
Contributor
Loading…
Use Python 3.10+ typing in core TBE ops (#5675)
cla signed
fb-exported
meta-exported
#5675
opened Apr 22, 2026 by
q10
Contributor
Loading…
Exclude transient RES streaming buffers from checkpoints by setting persistent=False
cla signed
fb-exported
meta-exported
#5674
opened Apr 22, 2026 by
FriedCosey
Loading…
Add FP8 rowwise padding to quantized AllToAll pooled embeddings
cla signed
fb-exported
meta-exported
#5673
opened Apr 22, 2026 by
RohanVardhan
Loading…
fbcode/deeplearning/fbgemm/fbgemm_gpu/test/tbe/training/merge_vbe_test.py
cla signed
fb-exported
meta-exported
#5672
opened Apr 22, 2026 by
meta-codesync
Bot
Loading…
Use Python 3.10+ typing in core TBE ops
cla signed
#5670
opened Apr 22, 2026 by
cyyever
Contributor
Loading…
Fix AMD build incompatibility and incorrect main_module paths
cla signed
fb-exported
meta-exported
#5668
opened Apr 21, 2026 by
q10
Contributor
Loading…
log query empty count vs total count
cla signed
fb-exported
meta-exported
#5657
opened Apr 17, 2026 by
xywang9334
Loading…
Fix VBE batch sizes not passed to request builder (#5653)
cla signed
fb-exported
meta-exported
#5653
opened Apr 17, 2026 by
gregmacnamara
Loading…
Port merge_embeddings benchmark to tritonbench
cla signed
fb-exported
meta-exported
#5650
opened Apr 16, 2026 by
q10
Contributor
Loading…
Validate total_num_blocks divisibility by my_size in block_bucketize (#5646)
cla signed
fb-exported
meta-exported
#5649
opened Apr 16, 2026 by
q10
Contributor
Loading…
Fix bf16 rounding to IEEE 754 ties-to-even
cla signed
#5648
opened Apr 16, 2026 by
cyyever
Contributor
Loading…
Add CPU support in fbgemm for FloatToFP8RowwiseQuantized and FP8RowwiseQuantizedToFloat (#5644)
cla signed
fb-exported
meta-exported
#5644
opened Apr 15, 2026 by
djjatmeta
Loading…
Fix TBE v2 forward kernel for embedding dim > 1024 (#5326) (#5569)
cla signed
fb-exported
meta-exported
#5641
opened Apr 15, 2026 by
q10
Contributor
Loading…
Add weight_init_device support for both Split and Dense TBE kernels
cla signed
fb-exported
meta-exported
#5640
opened Apr 15, 2026 by
TroyGarden
Contributor
Loading…
Fix intra-warp and inter-warp race conditions in bounds_check_indices v1 and v2 CUDA kernels
cla signed
fb-exported
meta-exported
#5638
opened Apr 15, 2026 by
gchalump
Contributor
Loading…
Add missing async proxy fence
cla signed
fb-exported
meta-exported
#5637
opened Apr 15, 2026 by
lw
Contributor
Loading…
Add aligned_unique_ptr RAII wrapper to avoid leak risks (#5609)
cla signed
fb-exported
meta-exported
#5615
opened Apr 11, 2026 by
q10
Contributor
Loading…
Port batched_dense_vec_jagged_2d_mul and jagged_1d_to_truncated_values to tritonbench
cla signed
fb-exported
meta-exported
#5603
opened Apr 9, 2026 by
q10
Contributor
Loading…
Replace rocm-smi with amd-smi across ROCm build, CI, and docs
cla signed
module: rocm
#5597
opened Apr 8, 2026 by
adam360x
Loading…
3 tasks done
bf16 scale/bias for INT4 (#5595)
cla signed
fb-exported
meta-exported
#5595
opened Apr 8, 2026 by
jeetkanjani7
Loading…
Enable more clang-tidy checks on C++20 (#5575)
cla signed
fb-exported
meta-exported
module: rocm
#5588
opened Apr 7, 2026 by
q10
Contributor
Loading…
Add gflag to select feature names for SSD KV embedding table
cla signed
fb-exported
meta-exported
#5585
opened Apr 7, 2026 by
jnwan
Loading…
Split RowWiseSparseAdagradFused.cc.stripped.o from fbcode//admarket/adfinder:adfinder
cla signed
fb-exported
meta-exported
#5578
opened Apr 6, 2026 by
meta-codesync
Bot
Loading…
Previous Next
ProTip!
What’s not been updated in a month: updated:<2026-03-23.