Skip to content

Actions: pytorch/FBGEMM

Actions

FBGEMM_GPU-CPU Benchmark

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
fbgemm_gpu_benchmark_cpu.yml
5,018 workflow runs
5,018 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Support bf16 in blackwell cutlass decode attention kernel FBGEMM_GPU-CPU Benchmark #5635: Pull request #4916 opened by Aya-ZIbra
Action required Aya-ZIbra:export-D82991495
Aya-ZIbra:export-D82991495
Action required
Optimize wgrad CUTLASS grouped gemm FBGEMM_GPU-CPU Benchmark #5634: Pull request #4891 synchronize by jiawenliu64
19m 0s export-D82700455
export-D82700455
19m 0s
Support multiple total-k and total-m in quantize bench FBGEMM_GPU-CPU Benchmark #5633: Pull request #4890 synchronize by jiawenliu64
20m 7s jiawenliu64:export-D82700396
jiawenliu64:export-D82700396
20m 7s
Support multiple total-k and total-m in quantize bench FBGEMM_GPU-CPU Benchmark #5632: Pull request #4890 synchronize by jiawenliu64
9m 57s jiawenliu64:export-D82700396
jiawenliu64:export-D82700396
9m 57s
Optimize wgrad CUTLASS grouped gemm FBGEMM_GPU-CPU Benchmark #5631: Pull request #4891 synchronize by jiawenliu64
16m 22s export-D82700455
export-D82700455
16m 22s
Support multiple total-k and total-m in quantize bench FBGEMM_GPU-CPU Benchmark #5630: Pull request #4890 synchronize by jiawenliu64
8m 5s jiawenliu64:export-D82700396
jiawenliu64:export-D82700396
8m 5s
Resolve wgrad grouped gemm relocation issue in fbgemm FBGEMM_GPU-CPU Benchmark #5629: Pull request #4915 opened by jiawenliu64
20m 30s export-D83020887
export-D83020887
20m 30s
Enable CUTLASS grouped GEMM for pretraining wgrad on GB200 and H100 (resubmit) FBGEMM_GPU-CPU Benchmark #5628: Pull request #4913 synchronize by jiawenliu64
20m 0s jiawenliu64:export-D83001505
jiawenliu64:export-D83001505
20m 0s
Optimize wgrad CUTLASS grouped gemm FBGEMM_GPU-CPU Benchmark #5627: Pull request #4891 synchronize by jiawenliu64
12m 16s export-D82700455
export-D82700455
12m 16s
[fbgemm_gpu] Fix IndicesEstimator constructor FBGEMM_GPU-CPU Benchmark #5626: Pull request #4906 synchronize by cyyever
20m 28s cyyever:IndicesEstimator_constructor
cyyever:IndicesEstimator_constructor
20m 28s
Migrate GenAI gqa attn splitk kernels to FBGEMM_LAUNCH_KERNEL, pt 2 FBGEMM_GPU-CPU Benchmark #5625: Pull request #4914 opened by q10
21m 9s q10:export-D81829067
q10:export-D81829067
21m 9s
Enable CUTLASS grouped GEMM for pretraining wgrad on GB200 and H100 (resubmit) FBGEMM_GPU-CPU Benchmark #5624: Pull request #4913 opened by jiawenliu64
10m 3s jiawenliu64:export-D83001505
jiawenliu64:export-D83001505
10m 3s
avx512 based int8 -> bf16 dequantization FBGEMM_GPU-CPU Benchmark #5623: Pull request #4912 synchronize by seanx92
17m 53s seanx92:export-D82507938
seanx92:export-D82507938
17m 53s
Add cutlass decode kernel to TritonBench FBGEMM_GPU-CPU Benchmark #5622: Pull request #4853 synchronize by Aya-ZIbra
19m 38s Aya-ZIbra:export-D80041532
Aya-ZIbra:export-D80041532
19m 38s
Add cutlass decode kernel to TritonBench FBGEMM_GPU-CPU Benchmark #5621: Pull request #4853 synchronize by Aya-ZIbra
12m 44s Aya-ZIbra:export-D80041532
Aya-ZIbra:export-D80041532
12m 44s
avx512 based int8 -> bf16 dequantization FBGEMM_GPU-CPU Benchmark #5620: Pull request #4912 opened by seanx92
22m 3s seanx92:export-D82507938
seanx92:export-D82507938
22m 3s
Disable SWA related decode kernel tests since they are not supported FBGEMM_GPU-CPU Benchmark #5619: Pull request #4911 opened by Alkaid-Benetnash
Action required Alkaid-Benetnash:export-D82991496
Alkaid-Benetnash:export-D82991496
Action required
Update to use Python 3.9 syntax (#4905) FBGEMM_GPU-CPU Benchmark #5618: Pull request #4909 synchronize by q10
21m 56s q10:export-D82979037
q10:export-D82979037
21m 56s
Update to use Python 3.9 syntax (#4905) FBGEMM_GPU-CPU Benchmark #5617: Pull request #4909 synchronize by q10
12m 39s q10:export-D82979037
q10:export-D82979037
12m 39s
Cut over coalesce ops FBGEMM_GPU-CPU Benchmark #5616: Pull request #4910 opened by cthi
24m 36s cthi:export-D82983166
cthi:export-D82983166
24m 36s
Update to use Python 3.9 syntax (#4905) FBGEMM_GPU-CPU Benchmark #5615: Pull request #4909 opened by q10
21m 53s q10:export-D82979037
q10:export-D82979037
21m 53s
Remove e5m2 from f8f8bf16_rowwise_batched FBGEMM_GPU-CPU Benchmark #5614: Pull request #4908 opened by cthi
19m 0s cthi:export-D82976652
cthi:export-D82976652
19m 0s
Remove e5m2 from f8f8bf16_rowwise FBGEMM_GPU-CPU Benchmark #5613: Pull request #4907 synchronize by cthi
19m 19s cthi:export-D82965586
cthi:export-D82965586
19m 19s
Remove e5m2 from f8f8bf16_rowwise FBGEMM_GPU-CPU Benchmark #5612: Pull request #4907 opened by cthi
21m 31s cthi:export-D82965586
cthi:export-D82965586
21m 31s
[fbgemm_gpu] Fix IndicesEstimator constructor FBGEMM_GPU-CPU Benchmark #5611: Pull request #4906 opened by cyyever
13m 26s cyyever:IndicesEstimator_constructor
cyyever:IndicesEstimator_constructor
13m 26s