Skip to content

Actions: q10/FBGEMM

Actions

FBGEMM CI

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
fbgemm_ci.yml
346 workflow runs
346 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Small modifications to quantize_bench script (#3684) FBGEMM CI #539: Commit 1b7789a pushed by q10
February 13, 2025 20:37 12m 33s main
main
February 13, 2025 20:37 12m 33s
refactor sweep_utils.py to test gemv kernel for different precisions … FBGEMM CI #538: Commit f3424d6 pushed by q10
February 12, 2025 21:23 15m 54s main
main
February 12, 2025 21:23 15m 54s
Support histogram_binning_calibration for export (#3657) FBGEMM CI #537: Commit d8e07ce pushed by q10
February 12, 2025 01:53 43m 58s main
main
February 12, 2025 01:53 43m 58s
Unifying TBE API using List (Backend) (#3563) FBGEMM CI #536: Commit fe80da4 pushed by q10
February 10, 2025 21:15 15m 50s main
main
February 10, 2025 21:15 15m 50s
Fix f8f8bf16_lite quantize op input in quantize_and_compute (#3667) FBGEMM CI #535: Commit a914871 pushed by q10
February 8, 2025 00:51 45m 1s main
main
February 8, 2025 00:51 45m 1s
loose unit test atol rtol tolerance to eliminate ut flakiness (#3… FBGEMM CI #534: Commit 9a343a0 pushed by q10
February 6, 2025 08:08 15m 10s main
main
February 6, 2025 08:08 15m 10s
fp8 rowwise regular gemm tuning for llm new shapes (#3654) FBGEMM CI #533: Commit bdcce9c pushed by q10
February 4, 2025 05:31 42m 0s main
main
February 4, 2025 05:31 42m 0s
Fix zero_start_index_M argument for triton rowwise quantize (#3639) FBGEMM CI #532: Commit 26eeef0 pushed by q10
February 3, 2025 19:01 50m 57s main
main
February 3, 2025 19:01 50m 57s
Use custom copy of cutlass to enable FP8 Grouped Gemm. (#3649) FBGEMM CI #531: Commit 0b2c24f pushed by q10
February 2, 2025 05:35 41m 49s main
main
February 2, 2025 05:35 41m 49s
Adding Missing includes and explicitly declaring Tensor in aten names… FBGEMM CI #530: Commit 3049ebe pushed by q10
January 30, 2025 23:57 12m 36s main
main
January 30, 2025 23:57 12m 36s
Add abstract impl for Fused8BitRowwiseQuantizedToFloatOrHalf et al. (… FBGEMM CI #529: Commit 8e79342 pushed by q10
January 30, 2025 19:35 20m 7s main
main
January 30, 2025 19:35 20m 7s
Add optional zero_start_index_M argument to triton fp8 rowwise quanti… FBGEMM CI #528: Commit 54e83db pushed by q10
January 29, 2025 18:16 18m 15s main
main
January 29, 2025 18:16 18m 15s
Minor doc fix (#3618) FBGEMM CI #527: Commit 73f6cee pushed by q10
January 28, 2025 19:05 49m 14s main
main
January 28, 2025 19:05 49m 14s
Move embedding_rocksdb_wrapper to its own header. (#3622) FBGEMM CI #526: Commit bab9b62 pushed by q10
January 28, 2025 07:06 42m 24s main
main
January 28, 2025 07:06 42m 24s
basic tbe input dump framework (#3593) FBGEMM CI #525: Commit e7c08e3 pushed by q10
January 27, 2025 18:10 44m 53s main
main
January 27, 2025 18:10 44m 53s
fix llm shapes in quantize bench and add ldm shapes (#3611) FBGEMM CI #524: Commit 1aff241 pushed by q10
January 24, 2025 21:17 12m 30s main
main
January 24, 2025 21:17 12m 30s
Updates and fixes to tensor_accessor.h (2I/N) (#3571) FBGEMM CI #523: Commit 4df8a12 pushed by q10
January 22, 2025 22:48 12m 8s main
main
January 22, 2025 22:48 12m 8s
Return if no data to allreduce (#3586) FBGEMM CI #522: Commit 6a684cb pushed by q10
January 22, 2025 18:25 45m 38s main
main
January 22, 2025 18:25 45m 38s
Back out "Enable fast FP8 GEMM for memory bound" (#3588) FBGEMM CI #521: Commit 2d025dc pushed by q10
January 19, 2025 21:59 13m 59s main
main
January 19, 2025 21:59 13m 59s
Fix specailization issue in keyed_jagged_index_select_dim1_forward_cu… FBGEMM CI #520: Commit 21d1260 pushed by q10
January 17, 2025 18:40 13m 54s main
main
January 17, 2025 18:40 13m 54s
more fp8 tuning for decode (#3576) FBGEMM CI #519: Commit 5c76d93 pushed by q10
January 16, 2025 18:50 13m 9s main
main
January 16, 2025 18:50 13m 9s
add patchelf as a required package in fbgemm_gpu/requirements.txt (#3… FBGEMM CI #518: Commit 9e9aa93 pushed by q10
January 16, 2025 00:16 12m 41s main
main
January 16, 2025 00:16 12m 41s
Add new optimizer state row_counter for Adam [Frontend] (#3558) FBGEMM CI #517: Commit 0783e80 pushed by q10
January 15, 2025 23:08 14m 23s main
main
January 15, 2025 23:08 14m 23s
Add build support for AMD MI300 (#3566) FBGEMM CI #516: Commit 06c88ad pushed by q10
January 14, 2025 01:42 18m 43s main
main
January 14, 2025 01:42 18m 43s
Add support for int32_t indices in TBE training (2D/N) (#3374) FBGEMM CI #515: Commit a93eed4 pushed by q10
January 13, 2025 18:13 16m 33s main
main
January 13, 2025 18:13 16m 33s