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
Fix optimized binary op broadcast rank mismatch Test XNNPACK Backend #11275: Pull request #21052 opened by Kilbex
Action required Kilbex:fix-optimized-binary-broadcast-rank-10959
Kilbex:fix-optimized-binary-broadcast-rank-10959
Action required
Record LLM model execution latency Test XNNPACK Backend #11274: Pull request #21040 synchronize by yenhao-huang
In progress yenhao-huang:fix/11771-llama-forward-latency
yenhao-huang:fix/11771-llama-forward-latency
In progress
[ExecuTorch][Vulkan] Add et_vk.linear_q4gsw_backward kernel (4-bit input-grad) Test XNNPACK Backend #11273: Pull request #20943 synchronize by JCNTH
46m 31s gh/JCNTH/82/head
gh/JCNTH/82/head
46m 31s
[ExecuTorch][Vulkan] Tests for et_vk.linear_dW Test XNNPACK Backend #11272: Pull request #20940 synchronize by JCNTH
1h 36m 18s gh/JCNTH/79/head
gh/JCNTH/79/head
1h 36m 18s
[ExecuTorch][Vulkan] Tests for et_vk.linear_q4gsw_backward Test XNNPACK Backend #11271: Pull request #20944 synchronize by JCNTH
46m 30s gh/JCNTH/83/head
gh/JCNTH/83/head
46m 30s
[ExecuTorch][Vulkan] Tests for et_vk.fused_ce Test XNNPACK Backend #11270: Pull request #20942 synchronize by JCNTH
46m 31s gh/JCNTH/81/head
gh/JCNTH/81/head
46m 31s
[ExecuTorch][Vulkan] Add et_vk.q4gsw_requant kernel (STE re-quant to W_4X8) Test XNNPACK Backend #11269: Pull request #20945 synchronize by JCNTH
46m 30s gh/JCNTH/84/head
gh/JCNTH/84/head
46m 30s
[ExecuTorch][Vulkan] Tests for et_vk.q4gsw_requant Test XNNPACK Backend #11268: Pull request #20946 synchronize by JCNTH
46m 32s gh/JCNTH/85/head
gh/JCNTH/85/head
46m 32s
[ExecuTorch][Vulkan] Tests for et_vk.adamw_step Test XNNPACK Backend #11267: Pull request #20938 synchronize by JCNTH
1h 35m 49s gh/JCNTH/77/head
gh/JCNTH/77/head
1h 35m 49s
[ExecuTorch][Vulkan] Add et_vk.linear_dW kernel (dense fp32 weight-grad GEMM) Test XNNPACK Backend #11266: Pull request #20939 synchronize by JCNTH
1h 35m 58s gh/JCNTH/78/head
gh/JCNTH/78/head
1h 35m 58s
[ExecuTorch][Vulkan] Add et_vk.fused_ce kernels (fused cross-entropy loss + dlogits) Test XNNPACK Backend #11265: Pull request #20941 synchronize by JCNTH
46m 33s gh/JCNTH/80/head
gh/JCNTH/80/head
46m 33s
[ExecuTorch][WebGPU] Glob runtime/ops sources in CMakeLists Test XNNPACK Backend #11264: Pull request #21005 synchronize by JCNTH
46m 32s gh/JCNTH/101/head
gh/JCNTH/101/head
46m 32s
Fix MODEL_CONVERTER_LIB_DIR path resolution for VGF tests (#20907) Test XNNPACK Backend #11263: Pull request #20907 synchronize by psiddh
3m 28s psiddh:export-D111819484
psiddh:export-D111819484
3m 28s
[ExecuTorch][WebGPU] Add optimized SDPA op Test XNNPACK Backend #11262: Pull request #20851 synchronize by JCNTH
2h 15m 16s gh/JCNTH/26/head
gh/JCNTH/26/head
2h 15m 16s
[ExecuTorch][WebGPU] Migrate 9 ops to make_compute_pipeline (layout:"auto") Test XNNPACK Backend #11261: Pull request #20868 synchronize by JCNTH
2h 15m 20s gh/JCNTH/43/head
gh/JCNTH/43/head
2h 15m 20s
[ExecuTorch][WebGPU] Add upsample_nearest2d op Test XNNPACK Backend #11260: Pull request #20859 synchronize by JCNTH
2h 15m 17s gh/JCNTH/34/head
gh/JCNTH/34/head
2h 15m 17s
[ExecuTorch][WebGPU] Add aten.sub.Tensor with TensorMeta broadcast Test XNNPACK Backend #11259: Pull request #20985 synchronize by JCNTH
2h 15m 6s gh/JCNTH/89/head
gh/JCNTH/89/head
2h 15m 6s
[ExecuTorch][WebGPU] et_vk.sdpa channel-attention tests Test XNNPACK Backend #11258: Pull request #20872 synchronize by JCNTH
2h 15m 0s gh/JCNTH/47/head
gh/JCNTH/47/head
2h 15m 0s
[ExecuTorch][WebGPU] Tests for aten._to_copy.default int<->float convert Test XNNPACK Backend #11257: Pull request #20988 synchronize by JCNTH
2h 12m 53s gh/JCNTH/92/head
gh/JCNTH/92/head
2h 12m 53s
[ExecuTorch][WebGPU] Add optimized addmm op Test XNNPACK Backend #11256: Pull request #20855 synchronize by JCNTH
2h 15m 9s gh/JCNTH/30/head
gh/JCNTH/30/head
2h 15m 9s
[ExecuTorch][WebGPU] linear_fp32 op tests Test XNNPACK Backend #11255: Pull request #20848 synchronize by JCNTH
16m 33s gh/JCNTH/23/head
gh/JCNTH/23/head
16m 33s
[ExecuTorch][WebGPU] Add optimized linear_fp32 op Test XNNPACK Backend #11254: Pull request #20847 synchronize by JCNTH
6s gh/JCNTH/22/head
gh/JCNTH/22/head
6s
[ExecuTorch][WebGPU] conv2d op tests Test XNNPACK Backend #11253: Pull request #20850 synchronize by JCNTH
6s gh/JCNTH/25/head
gh/JCNTH/25/head
6s
[ExecuTorch][WebGPU] Migrate multi-dispatch ops to make_compute_pipeline Test XNNPACK Backend #11252: Pull request #20869 synchronize by JCNTH
5s gh/JCNTH/44/head
gh/JCNTH/44/head
5s
[ExecuTorch][WebGPU] slice_copy Double-arg scalar handling + view_copy/alias glue Test XNNPACK Backend #11251: Pull request #20865 synchronize by JCNTH
10s gh/JCNTH/40/head
gh/JCNTH/40/head
10s