Skip to content

Actions: pytorch/executorch

Actions

Test XNNPACK Backend

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
test-backend-xnnpack.yml
2,500+ workflow runs
2,500+ workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
Add fused GatedDeltaNet decode Triton kernel Test XNNPACK Backend #2516: Pull request #18865 synchronize by Gasoonjia
56m 25s fused-deltanet-decode
fused-deltanet-decode
56m 25s
try to use KA optimize moe kernel Test XNNPACK Backend #2515: Pull request #18930 synchronize by Gasoonjia
1m 5s kamoe
kamoe
1m 5s
Make quantized_conv1d_depthwise kernels call the 2d optimized version… Test XNNPACK Backend #2514: Commit 4b20342 pushed by meta-codesync Bot
1h 1m 4s main
main
1h 1m 4s
Implement FlashDecoding++ async softmax for split-K SDPA Test XNNPACK Backend #2513: Pull request #18867 synchronize by Gasoonjia
1h 4m 30s gasoonjia/flashdecoding-pp-async-softmax
gasoonjia/flashdecoding-pp-async-softmax
1h 4m 30s
Implement FlashDecoding++ async softmax for split-K SDPA Test XNNPACK Backend #2512: Pull request #18867 synchronize by Gasoonjia
3m 51s gasoonjia/flashdecoding-pp-async-softmax
gasoonjia/flashdecoding-pp-async-softmax
3m 51s
Split regular and depthwise quantized conv1d (#18807) Test XNNPACK Backend #2511: Commit 1fc75d0 pushed by meta-codesync Bot
1h 0m 1s main
main
1h 0m 1s
Add arm-vgf-library v0.9.0 and update vgf_backend dep Test XNNPACK Backend #2510: Commit f5ccb62 pushed by meta-codesync Bot
1h 0m 0s main
main
1h 0m 0s
Add fused GatedDeltaNet decode Triton kernel Test XNNPACK Backend #2509: Pull request #18865 synchronize by Gasoonjia
1h 0m 33s fused-deltanet-decode
fused-deltanet-decode
1h 0m 33s
Add MLX op handler for aten.isinf Test XNNPACK Backend #2508: Pull request #18936 opened by Ai-chan-0411
1h 5m 55s Ai-chan-0411:fix/mlx-isinf-handler
Ai-chan-0411:fix/mlx-isinf-handler
1h 5m 55s
Fix #18562: Method.execute() silently produces wrong results for no... Test XNNPACK Backend #2507: Pull request #18935 opened by JiwaniZakir
57m 7s JiwaniZakir:fix/18562-method-execute-silently-produces-wrong
JiwaniZakir:fix/18562-method-execute-silently-produces-wrong
57m 7s
[aoti-cuda] fuse MoE fixes Test XNNPACK Backend #2506: Pull request #18934 opened by digantdesai
57m 38s digantdesai/fused_moe_fixes
digantdesai/fused_moe_fixes
57m 38s
Add arm-vgf-library v0.9.0 and update vgf_backend dep Test XNNPACK Backend #2505: Pull request #18932 opened by digantdesai
59m 26s digantdesai:export-D101069501
digantdesai:export-D101069501
59m 26s
[pytorch hash update] update the pinned pytorch hash Test XNNPACK Backend #2504: Pull request #10955 synchronize by pytorchupdatebot
1h 56m 37s update-pytorch-commit-hash/15080763801-537-1
update-pytorch-commit-hash/15080763801-537-1
1h 56m 37s
Test XNNPACK Backend Test XNNPACK Backend #2503: Scheduled
1h 9m 16s main
main
1h 9m 16s
Add multi-reader tests for Add/Sub rescale fusion (#18758) Test XNNPACK Backend #2502: Pull request #18758 synchronize by 3l1
55m 41s export-D99939008
export-D99939008
55m 41s
Add multi-reader tests for Add/Sub rescale fusion (#18758) Test XNNPACK Backend #2501: Pull request #18758 synchronize by meta-codesync Bot
4m 57s export-D99939008
export-D99939008
4m 57s
Arm backend: Disable model evaluation buck test Test XNNPACK Backend #2500: Commit ec8d70b pushed by meta-codesync Bot
32m 9s main
main
32m 9s
Support whole-number float exponents in ReplacePowWithMulPass (#18851) Test XNNPACK Backend #2499: Commit a14c946 pushed by meta-codesync Bot
1h 7m 47s main
main
1h 7m 47s
Fix dependency cycle in GroupBasedPartitioner._can_merge_partitions Test XNNPACK Backend #2498: Pull request #18397 synchronize by Hyungkeun-Park-Nota
57m 13s Hyungkeun-Park-Nota:fix/group-partitioner-cycle-detection
Hyungkeun-Park-Nota:fix/group-partitioner-cycle-detection
57m 13s
Add MLX op handler for aten.bitwise_xor Test XNNPACK Backend #2497: Pull request #18931 opened by nanookclaw
Action required nanookclaw:mlx-bitwise-xor
nanookclaw:mlx-bitwise-xor
Action required
Make quantized_conv1d_depthwise kernels call the 2d optimized versions on HiFi (#18883) Test XNNPACK Backend #2496: Pull request #18883 synchronize by mcremon-meta
57m 38s export-D99731352
export-D99731352
57m 38s
try to use KA optimize moe kernel Test XNNPACK Backend #2495: Pull request #18930 opened by Gasoonjia
1h 3m 51s kamoe
kamoe
1h 3m 51s
Make quantized_conv1d_depthwise kernels call the 2d optimized versions on HiFi (#18883) Test XNNPACK Backend #2494: Pull request #18883 synchronize by meta-codesync Bot
2m 50s export-D99731352
export-D99731352
2m 50s
Split regular and depthwise quantized conv1d (#18807) Test XNNPACK Backend #2493: Pull request #18807 synchronize by meta-codesync Bot
1h 0m 4s export-D99509365
export-D99509365
1h 0m 4s
Interface for swapping inputs and outputs based on cost model + working permute example (#18929) Test XNNPACK Backend #2492: Pull request #18929 opened by DrJessop
1h 2m 58s DrJessop:export-D100917820
DrJessop:export-D100917820
1h 2m 58s
ProTip! You can narrow down the results and go further in time using created:<2026-04-16 or the other filters available.