Skip to content

Pull requests: pytorch/FBGEMM

New pull request New
576 Open 4,294 Closed
576 Open 4,294 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

Use int64_t instead of int32_t in decode attention kernel to avoid index calc overflow cla signed fb-exported meta-exported
#5095 opened Nov 5, 2025 by Alkaid-Benetnash Loading…
2
Deprecate tl.async_task from fbgemm cla signed fb-exported meta-exported
#5094 opened Nov 5, 2025 by dshi7 Loading…
2
Merge VBE output [backend] reland ci-no-td cla signed fb-exported meta-exported
#5093 opened Nov 5, 2025 by spcyppt Loading…
2
Bm/cuda 13e cla signed
#5090 opened Nov 5, 2025 by q10 Loading…
1
Fix NAN for the prediction (#2096) cla signed fb-exported meta-exported
#5088 opened Nov 4, 2025 by quhang Loading…
2
Support larger lookup in permute (#2094) cla signed fb-exported meta-exported
#5086 opened Nov 4, 2025 by kausv Loading…
2
Change kvzch_eviction_tbe_config to kvzch_tbe_config cla signed fb-exported meta-exported
#5084 opened Nov 4, 2025 by EddyLXJ Loading…
2
optimization: move set_metadata to ssd_eviction_stream cla signed fb-exported meta-exported
#5082 opened Nov 4, 2025 by emlin Loading…
2
Back out "bucket_permute kernel to accelerate bucktize in rebatching" cla signed fb-exported meta-exported
#5079 opened Nov 3, 2025 by royren622 Loading…
2
Several kernel optimization from aiter team cla signed module: rocm
#5074 opened Oct 31, 2025 by Bernard-Liu • Draft
4
log all table names in TBE cla signed fb-exported meta-exported
#5071 opened Oct 30, 2025 by ashuaibi7 Loading…
1
Efficient Causal/Local scheduling cla signed fb-exported meta-exported
#5066 opened Oct 28, 2025 by Aya-ZIbra Loading…
2
embedding forward optimization for MI350 cla signed
#5064 opened Oct 28, 2025 by JaxChen29 Loading…
4
Kernel interface for use in xformers cla signed fb-exported meta-exported
#5063 opened Oct 28, 2025 by Aya-ZIbra Loading…
2
Adding Kineto support to bench:sparse_ops cla signed fb-exported meta-exported
#5060 opened Oct 27, 2025 by gchalump Loading…
2
Support no eviction in Feature score eviction policy cla signed fb-exported meta-exported
#5059 opened Oct 27, 2025 by EddyLXJ Loading…
2
Abstract out sharable interface of Dram KV wrapper cla signed fb-exported meta-exported
#5057 opened Oct 27, 2025 by tomlintbl Loading…
2
Changing Backend Tensor initialization cla signed fb-exported meta-exported
#5056 opened Oct 25, 2025 by Raahul46 Loading…
2
Changing Backend Tensor initialization cla signed fb-exported meta-exported
#5055 opened Oct 25, 2025 by Raahul46 Loading…
2
Update embedding_forward_quantized_cpu_template.cpp to use initialized output memory instead of uninitialized cla signed fb-exported meta-exported
#5054 opened Oct 25, 2025 by quhang Loading…
2
use reorder_batched_ad_indices_vec kernel for rebatching cla signed fb-exported meta-exported
#5049 opened Oct 24, 2025 by royren622 Loading…
1
Remove AVX compilation on aarch64 builds cla signed fb-exported meta-exported
#5045 opened Oct 23, 2025 by Nicoshev Loading…
2
fix potential overflow will multiple attempts (#2054) cla signed fb-exported meta-exported
#5044 opened Oct 23, 2025 by emlin Loading…
2
PyTorch version compatibility check in _load_library cla signed fb-exported meta-exported
#5041 opened Oct 21, 2025 by gchalump Loading…
2
Fix build CI bug for release cla signed fb-exported meta-exported
#5035 opened Oct 21, 2025 by spcyppt Loading…
2
ProTip! no:milestone will show everything without a milestone.