Skip to content

Pull requests: pytorch/FBGEMM

New pull request New
465 Open 3,543 Closed
465 Open 3,543 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

Track updated rows in SSDTBE cla signed fb-exported
#4211 opened May 29, 2025 by chouxi Loading…
2
Pass the updated embeddings to EmbeddingKVDB cla signed fb-exported
#4210 opened May 29, 2025 by chouxi Loading…
2
Optimize cudaGetDeviceProperties runtime overhead cla signed fb-exported
#4209 opened May 29, 2025 by jiawenliu64 Loading…
5
Limit the grid size for the TBE forward kernel cla signed fb-exported
#4208 opened May 29, 2025 by q10 Loading…
2
Adding a mutex lock to set_range function cla signed fb-exported
#4207 opened May 29, 2025 by Raahul46 Loading…
2
Add HSTU in fbgemm_gpu/experimental/ (#4090) cla signed fb-exported
#4205 opened May 28, 2025 by ionuthristodorescu Loading…
2
Fused SILU with quantization and RMS with quantization cla signed fb-exported
#4204 opened May 28, 2025 by Tianyu-Liang Loading…
2
Support BF16 in Triton downcast quantization mx4 unpack kernel cla signed fb-exported
#4203 opened May 28, 2025 by Tianyu-Liang Loading…
2
add more checks for CPU combined inputs coalescing cla signed fb-exported
#4202 opened May 28, 2025 by 842974287 Loading…
2
Update the rowwise adagrad optimizer to leverage optimizer state offloading, v4, backend cla signed fb-exported
#4195 opened May 27, 2025 by q10 Loading…
6
Migrate TBE UVM cache kernels to FBGEMM_LAUNCH_KERNEL cla signed fb-exported
#4193 opened May 27, 2025 by q10 Loading…
2
WeightRow cleanups cla signed fb-exported
#4189 opened May 26, 2025 by q10 Loading…
2
feature evict cla signed
#4187 opened May 26, 2025 by ArronHZG Loading…
2
Add HSTU in fbgemm_gpu/experimental/ (#4090) cla signed
#4183 opened May 23, 2025 by ionuthristodorescu Loading…
1
Fixed the HSTU lint errors cla signed
#4182 opened May 23, 2025 by ionuthristodorescu Loading…
1
FP4 Triton kernel bug fix cla signed fb-exported
#4181 opened May 23, 2025 by Tianyu-Liang Loading…
4
Decouple some operator defs from operator impl cla signed fb-exported
#4175 opened May 22, 2025 by PatriceVignola Loading…
15
Add best config for FP8 medium shapes cla signed fb-exported
#4169 opened May 21, 2025 by njriasan Loading…
4
Update matmul_fp8_row_meta signature cla signed fb-exported
#4168 opened May 21, 2025 by njriasan Loading…
4
Support Bias in _kernel_matmul_fp8_row_non_persistent cla signed fb-exported
#4167 opened May 21, 2025 by njriasan Loading…
4
Update default persistence for matmul_fp8_row cla signed fb-exported
#4166 opened May 21, 2025 by njriasan Loading…
2
Jemalloc Mempool and Adaptation for CPU HASHTABLE cla signed
#4154 opened May 20, 2025 by ArronHZG Loading…
3
improve read/write performance by 100% cla signed fb-exported
#4150 opened May 19, 2025 by steven1327 Loading…
2
Make iter persistent for AdagradW cla signed fb-exported
#4147 opened May 17, 2025 by minhua-chen Loading…
5
reduce UT emb table size cla signed fb-exported
#4136 opened May 16, 2025 by duduyi2013 Loading…
2
ProTip! no:milestone will show everything without a milestone.