Skip to content

Pull requests: pytorch/FBGEMM

New pull request New
442 Open 3,426 Closed
442 Open 3,426 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

Add external qparams parameters to dequantize_int4_cache API cla signed fb-exported
#4062 opened May 1, 2025 by PatriceVignola Loading…
2
Allow merge_pooled_embedding take in device without index cla signed fb-exported
#4061 opened May 1, 2025 by ZhengkaiZ Loading…
7
Integrate compat with Trunk and enable disabling BufferOps cla signed fb-exported
#4060 opened May 1, 2025 by njriasan Loading…
2
fbgemm_gpu.experimental.gen_ai.moe.silu_mul_quant. cla signed fb-exported
#4059 opened Apr 30, 2025 by levendlee Loading…
12
Add logic to stream weights in EmbeddingKVDB cla signed fb-exported
#4058 opened Apr 30, 2025 by chouxi Loading…
2
Add enable_raw_embedding_streaming from TBE config to EmbeddingKVDB cla signed fb-exported
#4053 opened Apr 30, 2025 by chouxi Loading…
4
Simplify weight row cache load and evict routines cla signed fb-exported
#4050 opened Apr 30, 2025 by q10 Loading…
2
Refactor bounds_check_indices cla signed fb-exported
#4049 opened Apr 30, 2025 by sryap Loading…
3
Add more parameter specializations for autovec TBE kernels cla signed fb-exported
#4047 opened Apr 30, 2025 by excelle08 Loading…
4
ReportTBE data configuration with EEG-based indices (squash stack from D73450767) cla signed fb-exported
#4046 opened Apr 30, 2025 by gchalump Loading…
6
Fix int32_t to auto for code around WeightRow cla signed fb-exported
#4045 opened Apr 30, 2025 by q10 Loading…
2
integrate dramKV with kvtensorwrapper cla signed fb-exported
#4043 opened Apr 29, 2025 by steven1327 Loading…
2
torchrec support on kvzch emb lookup module cla signed fb-exported
#4035 opened Apr 28, 2025 by duduyi2013 Loading…
2
support zero collision tables in ssd operator cla signed fb-exported
#4033 opened Apr 28, 2025 by duduyi2013 Loading…
2
Enable NaN checks on tensor arguments to kernel launches cla signed fb-exported
#4029 opened Apr 26, 2025 by q10 Loading…
10
update hipify_torch submodule for version 2 cla signed
#4028 opened Apr 26, 2025 by jeffdaily Loading…
1
Add keep_orig_idx_per_feature parameter to block_bucketize_sparse_features kernel cla signed fb-exported
#4027 opened Apr 25, 2025 by emlin Loading…
6
Migrate make_pta_acc_format() away from old macros, v2] cla signed fb-exported
#4026 opened Apr 25, 2025 by q10 Loading…
2
Clean up WeightRow in preparation for optimizer state offloading cla signed fb-exported
#4021 opened Apr 24, 2025 by q10 Loading…
23
fix build that excludes a bunch of features cla signed fb-exported
#4019 opened Apr 24, 2025 by q10 Loading…
2
Report TBE data configuration with EEG-based indices estimation cla signed fb-exported
#4018 opened Apr 24, 2025 by gchalump Loading…
10
Gen modes: Remove -Wno-mismatched-tags cla signed fb-exported
#4011 opened Apr 23, 2025 by q10 Loading…
2
Back out "Add a workaround for stochastic rounding for AMD GPUs" cla signed fb-exported module: rocm
#3977 opened Apr 17, 2025 by ionuthristodorescu Loading…
2
Back out "Cleanups to StochasticRoundingRNGState" cla signed fb-exported module: rocm
#3976 opened Apr 17, 2025 by ionuthristodorescu Loading…
2
Remove C10_DEPRECATED cla signed fb-exported
#3960 opened Apr 11, 2025 by desertfire Loading…
2
ProTip! Filter pull requests by the default branch with base:main.