Skip to content

Pull requests: pytorch/FBGEMM

New pull request New
488 Open 3,647 Closed
488 Open 3,647 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

add ckpt and restore with feature evict metaheader cla signed
#4342 opened Jun 13, 2025 by lalala-2 Loading…
5
Fixing reading from EmbeddingRocksDB connection cla signed fb-exported
#4341 opened Jun 13, 2025 by Raahul46 Loading…
2
Making create_rocksdb_hard_link_snapshot function a no_op cla signed fb-exported
#4340 opened Jun 13, 2025 by Raahul46 Loading…
2
Implement a stat library for fbgemm embedding cla signed fb-exported
#4339 opened Jun 13, 2025 by Kaiweitu Loading…
2
[fbgemm_gpu] Upgrade benchmark workflows cla signed module: rocm
#4337 opened Jun 12, 2025 by q10 Loading…
1
Add initial version of TuningCache and scripts for heuristic + kernel (#4289) cla signed fb-exported
#4336 opened Jun 12, 2025 by cthi Loading…
5
Adding a separate utils file for KVTensorMetaData (#4298) cla signed fb-exported
#4335 opened Jun 12, 2025 by Raahul46 Loading…
4
Fix int_nbit inference int8 nobag kernel meta function cla signed fb-exported
#4333 opened Jun 12, 2025 by spcyppt Loading…
2
Try code comment-out hack cla signed fb-exported
#4331 opened Jun 12, 2025 by q10 Loading…
2
TBE benchmarks cla signed fb-exported
#4330 opened Jun 12, 2025 by q10 Loading…
2
Tune FP8 grouped GEMM for Llama4 shapes cla signed fb-exported
#4326 opened Jun 11, 2025 by jiawenliu64 Loading…
2
fix output dtype issue in merge_pooled_embeddings when input tensors are all empty cla signed fb-exported
#4325 opened Jun 11, 2025 by 842974287 Loading…
2
NVFP4 quantization emulation kernels as reference cla signed fb-exported
#4324 opened Jun 11, 2025 by summerdengfb Loading…
7
Use local counter for TBE boundary check warinings to improve performance cla signed fb-exported
#4316 opened Jun 10, 2025 by yoyoyocmu Loading…
2
Support prefetch pipeline in bounds_check_indices cla signed fb-exported
#4312 opened Jun 9, 2025 by sryap Loading…
2
Improve heuristic for Cutlass FP8 Grouped GEMM cla signed fb-exported
#4309 opened Jun 9, 2025 by cthi Loading…
11
Support tuning cache for Cutlass FP8 Grouped GEMM cla signed fb-exported
#4308 opened Jun 9, 2025 by cthi Loading…
9
[fbgemm_gpu] TBE microbenchmark upgrades cla signed module: rocm
#4307 opened Jun 9, 2025 by q10 Loading…
1
patch fixes for eviction cla signed fb-exported
#4304 opened Jun 9, 2025 by duduyi2013 Loading…
10
tbe cpu nobag dispatch and backward pass kernel impl cla signed fb-exported
#4303 opened Jun 9, 2025 by yabalaban Loading…
14
tbe cpu nobag dispatch and forward pass kernel impl cla signed fb-exported
#4302 opened Jun 9, 2025 by yabalaban Loading…
12
Support tuning cache for Cutlass FP8 GEMM cla signed fb-exported
#4301 opened Jun 9, 2025 by cthi Loading…
10
Add new kernels for Cutlass BF16 grouped GEMM for tuning cache cla signed fb-exported
#4300 opened Jun 9, 2025 by cthi Loading…
10
Support tuning cache for Cutlass BF16 grouped GEMM cla signed fb-exported
#4299 opened Jun 9, 2025 by cthi Loading…
8
put feature_evict definition in cpp file cla signed fb-exported
#4294 opened Jun 8, 2025 by chenyuzhcy Loading…
2
ProTip! Add no:assignee to see everything that’s not assigned.