Skip to content

Pull requests: pytorch/FBGEMM

New pull request New
478 Open 3,598 Closed
478 Open 3,598 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

Fix Signed-Unsigned Comparison in Tensor Utils cla signed fb-exported
#4279 opened Jun 6, 2025 by wenxin0319 Loading…
4
Migrate jagged tensor kernels to FBGEMM_LAUNCH_KERNEL, pt 1 cla signed fb-exported
#4278 opened Jun 5, 2025 by q10 Loading…
2
improve write performance by ~10x cla signed fb-exported
#4277 opened Jun 5, 2025 by steven1327 Loading…
2
Update AI Codesign Cutlass to 4.0 cla signed fb-exported
#4276 opened Jun 5, 2025 by jwfromm Loading…
3
Added triton implementation for nvfp4 quantization scheme cla signed fb-exported
#4275 opened Jun 5, 2025 by Tianyu-Liang Loading…
2
Deprecate float32 bias for Cutlass FP8 rowwise cla signed fb-exported
#4274 opened Jun 5, 2025 by cthi Loading…
2
OSS the murmur_hash function cla signed fb-exported
#4273 opened Jun 5, 2025 by lizhouyu Loading…
3
Decouple some operator defs from operator impl cla signed fb-exported
#4272 opened Jun 5, 2025 by PatriceVignola Loading…
2
Skip ROCm test for MPZCH kernel cla signed fb-exported module: rocm
#4271 opened Jun 5, 2025 by lizhouyu Loading…
2
skip ROCm tests cla signed fb-exported module: rocm
#4270 opened Jun 5, 2025 by lizhouyu Loading…
2
Attempt to fix the operator [] error in CUDA Kernel cla signed fb-exported
#4269 opened Jun 5, 2025 by lizhouyu Loading…
3
Attempts to solve errors caused by undefined function "CHECK_NOTNULL" - Use TORCH_CHECK instead cla signed fb-exported
#4266 opened Jun 4, 2025 by lizhouyu Loading…
4
Attempts to solve errors caused by undefined function "CHECK_NOTNULL" - Move common header back to src folder cla signed fb-exported
#4265 opened Jun 4, 2025 by lizhouyu Loading…
2
Vectorize load/store for get_FP8_qparam_cuda_kernel cla signed fb-exported
#4263 opened Jun 4, 2025 by flaviotruzzi Loading…
6
Vectorize load/store for FP8 Quantization cla signed fb-exported
#4262 opened Jun 4, 2025 by flaviotruzzi Loading…
21
Clean up tensor accessor tests cla signed fb-exported
#4258 opened Jun 4, 2025 by q10 Loading…
3
Add headers and modify the cmake cla signed fb-exported
#4256 opened Jun 4, 2025 by lizhouyu Loading…
5
HSTU CI v2 cla signed
#4251 opened Jun 4, 2025 by q10 Loading…
1
Update the rowwise adagrad optimizer to leverage optimizer state offloading, v4, frontend cla signed fb-exported
#4249 opened Jun 3, 2025 by q10 Loading…
5
Test outside of namespace reference cla signed fb-exported
#4248 opened Jun 3, 2025 by aporialiao Loading…
2
Revert D75037710: Decouple some operator defs from operator impl ci-no-td cla signed fb-exported
#4244 opened Jun 3, 2025 by PatriceVignola Loading…
2
HSTU CI v1 cla signed
#4236 opened Jun 2, 2025 by q10 Loading…
1
CK Preshuffle FP8 Rowwise GEMM cla signed fb-exported
#4234 opened Jun 1, 2025 by jwfromm Loading…
12
Simplify CK FP8 Kernel Launch and enable FP16 Outputs. cla signed fb-exported
#4233 opened Jun 1, 2025 by jwfromm Loading…
9
Added unit tests for the entire ssd offloading using rocksdb checkpoint flow cla signed fb-exported
#4228 opened May 30, 2025 by Raahul46 Loading…
3
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.