Skip to content

Actions: pytorch/executorch

Actions

Test WebGPU Backend

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
test-backend-webgpu.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
[MLX] Add quant regrouping to MLX lowering flow, and update quantize_and_save in Qwen to support MLX path Test WebGPU Backend #3876: Pull request #21023 synchronize by metascroy
In progress mlx-quant-improvements
mlx-quant-improvements
In progress
Reland "Arm backend: Add transpose propagation pass" v2 Test WebGPU Backend #3875: Pull request #21030 opened by rascani
In progress revert-20947-revert-20748-revert-20742-revert-20625-change-1265162
revert-20947-revert-20748-revert-20742-revert-20625-change-1265162
In progress
Revert "Reland "Arm backend: Add transpose propagation pass" (#20748)" Test WebGPU Backend #3874: Commit 8077bb6 pushed by meta-codesync Bot
In progress main
main
In progress
Reland "Arm backend: Make composable_quantizer default" Test WebGPU Backend #3873: Commit 7bb34e1 pushed by meta-codesync Bot
In progress main
main
In progress
Qualcomm AI Engine Direct - Adding QNN backend support for scatter.value core ATen op Test WebGPU Backend #3872: Pull request #20998 synchronize by qti-horodnic
Action required CodeLinaro:scatter_value
CodeLinaro:scatter_value
Action required
Fix SourceAv in function 'flatbuffers::ReadScalar' Test WebGPU Backend #3871: Pull request #21029 opened by rascani
In progress export-D112525269
export-D112525269
In progress
Update excluded authors list in OSS contrib workflow (#21017) Test WebGPU Backend #3870: Commit abdc92d pushed by nil-is-all
In progress main
main
In progress
Cortex-M: quantize aten.matmul by rewriting it to bmm before annotation Test WebGPU Backend #3869: Pull request #21028 opened by rascani
59m 50s rascani:cortexm-matmul-bmm-pr
rascani:cortexm-matmul-bmm-pr
59m 50s
Cortex-M: fold constant attention-score scale into the softmax-input quantize Test WebGPU Backend #3868: Pull request #21027 opened by rascani
1h 3m 9s rascani:cortexm-scale-fold
rascani:cortexm-scale-fold
1h 3m 9s
Remove 16 redundant skipped test stubs in test_scalars.py Test WebGPU Backend #3867: Pull request #21026 opened by psiddh
1h 6m 44s psiddh:export-D112509983
psiddh:export-D112509983
1h 6m 44s
Remove _use_edge_ops=False (#20909) Test WebGPU Backend #3866: Pull request #20909 synchronize by JakeStevens
57m 36s JakeStevens:export-D111926304
JakeStevens:export-D111926304
57m 36s
Reorder channelwise gated delta rule chunked hot loops for autovectorization (#21021) Test WebGPU Backend #3865: Pull request #21021 synchronize by JakeStevens
1h 6m 48s JakeStevens:export-D112598714
JakeStevens:export-D112598714
1h 6m 48s
Two-pass optimization for channelwise gated delta rule kernel (#21020) Test WebGPU Backend #3864: Pull request #21020 synchronize by JakeStevens
1h 17m 2s JakeStevens:export-D112596724
JakeStevens:export-D112596724
1h 17m 2s
Cortex-M backend: lower quantized GELU via the activation LUT Test WebGPU Backend #3863: Pull request #21025 opened by rascani
1h 1m 25s rascani:cortexm-gelu-lut
rascani:cortexm-gelu-lut
1h 1m 25s
Add fused int8 LSTM support to the Cortex-M (CMSIS-NN) backend Test WebGPU Backend #3862: Pull request #21024 opened by rascani
1h 0m 58s rascani:cortex-m-lstm
rascani:cortex-m-lstm
1h 0m 58s
[MLX] Add quant regrouping to MLX lowering flow, and update quantize_and_save in Qwen to support MLX path Test WebGPU Backend #3861: Pull request #21023 opened by metascroy
58m 28s mlx-quant-improvements
mlx-quant-improvements
58m 28s
Two-pass optimization for channelwise gated delta rule kernel (#21020) Test WebGPU Backend #3860: Pull request #21020 synchronize by JakeStevens
43m 52s JakeStevens:export-D112596724
JakeStevens:export-D112596724
43m 52s
Gate output-centric image_to_nchw staging to discrete GPUs Test WebGPU Backend #3859: Pull request #21022 opened by SS-JIA
1h 8m 45s SS-JIA:export-D112599710
SS-JIA:export-D112599710
1h 8m 45s
Reorder channelwise gated delta rule chunked hot loops for autovectorization (#21021) Test WebGPU Backend #3858: Pull request #21021 opened by JakeStevens
18m 4s JakeStevens:export-D112598714
JakeStevens:export-D112598714
18m 4s
Two-pass optimization for channelwise gated delta rule kernel (#21020) Test WebGPU Backend #3857: Pull request #21020 opened by JakeStevens
6m 26s JakeStevens:export-D112596724
JakeStevens:export-D112596724
6m 26s
Add channel-wise gated delta rule custom op (#20913) Test WebGPU Backend #3856: Commit 931ba24 pushed by meta-codesync Bot
1h 2m 41s main
main
1h 2m 41s
Remove _use_edge_ops=False (#20909) Test WebGPU Backend #3855: Pull request #20909 synchronize by JakeStevens
59m 19s JakeStevens:export-D111926304
JakeStevens:export-D111926304
59m 19s
Fix portable pybindings COW ABI dependency Test WebGPU Backend #3854: Pull request #21019 opened by JakeStevens
49m 36s JakeStevens:fix-portable-pybindings-cow-abi
JakeStevens:fix-portable-pybindings-cow-abi
49m 36s
Revert "Reland "Arm backend: Add transpose propagation pass" (#20748)" Test WebGPU Backend #3853: Pull request #20947 synchronize by rascani
1h 7m 3s revert-20748-revert-20742-revert-20625-change-1265162
revert-20748-revert-20742-revert-20625-change-1265162
1h 7m 3s
Arm backend: Add DeepSeek-R1-Distill-Qwen FP8 layer tests Test WebGPU Backend #3852: Pull request #21012 synchronize by bdemirb
1h 3m 57s bdemirb:baris_mletorch-2337_fp8_layer_test_DeepSeek_R1_Distill_Qwen_1.5B
bdemirb:baris_mletorch-2337_fp8_layer_test_DeepSeek_R1_Distill_Qwen_1.5B
1h 3m 57s