Skip to content

Actions: pytorch/executorch

Actions

Test WebGPU Backend

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
test-backend-webgpu.yml
1,620 workflow runs
1,620 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 pthreadpool subset corruption + executor deadlock; de-nest ExecuT… Test WebGPU Backend #1403: Commit 68f0768 pushed by meta-codesync Bot
52m 43s main
main
52m 43s
Arm backend: Fix test_resnet18.py for Arm64 (#20335) Test WebGPU Backend #1402: Commit 15b0d73 pushed by zingo
58m 5s main
main
58m 5s
Arm backend: Lower MXFP Linear to TOSA (#20065) Test WebGPU Backend #1401: Commit 6dab6b1 pushed by zingo
1h 0m 27s main
main
1h 0m 27s
Add a weight-stationary INT4 GEMM for small M Test WebGPU Backend #1400: Pull request #20154 synchronize by digantdesai
1h 1m 46s gh/digantdesai/59/head
gh/digantdesai/59/head
1h 1m 46s
Add KV cache to the EAGLE-3 draft head Test WebGPU Backend #1399: Pull request #20152 synchronize by digantdesai
1h 2m 16s gh/digantdesai/57/head
gh/digantdesai/57/head
1h 2m 16s
Add the EAGLE-3 speculator CUDA export Test WebGPU Backend #1398: Pull request #20155 synchronize by digantdesai
1h 2m 2s gh/digantdesai/60/head
gh/digantdesai/60/head
1h 2m 2s
Add EAGLE-3 end-to-end speculative-decode test Test WebGPU Backend #1397: Pull request #20157 synchronize by digantdesai
1h 1m 41s gh/digantdesai/62/head
gh/digantdesai/62/head
1h 1m 41s
Add KV cache to the EAGLE-3 draft head Test WebGPU Backend #1396: Pull request #20152 synchronize by digantdesai
3s gh/digantdesai/57/head
gh/digantdesai/57/head
3s
Add the Eagle3Speculator module for speculative decoding Test WebGPU Backend #1395: Pull request #20153 synchronize by digantdesai
1h 2m 23s gh/digantdesai/58/head
gh/digantdesai/58/head
1h 2m 23s
Add the EAGLE-3 speculative-decoding runner (CUDA) Test WebGPU Backend #1394: Pull request #20156 synchronize by digantdesai
42m 23s gh/digantdesai/61/head
gh/digantdesai/61/head
42m 23s
Add the Eagle3Speculator module for speculative decoding Test WebGPU Backend #1393: Pull request #20153 synchronize by digantdesai
7s gh/digantdesai/58/head
gh/digantdesai/58/head
7s
Add the EAGLE-3 speculator CUDA export Test WebGPU Backend #1392: Pull request #20155 synchronize by digantdesai
10s gh/digantdesai/60/head
gh/digantdesai/60/head
10s
Add Gemma4 EAGLE-3 eager reference Test WebGPU Backend #1391: Pull request #20151 synchronize by digantdesai
49m 14s gh/digantdesai/56/head
gh/digantdesai/56/head
49m 14s
Add a weight-stationary INT4 GEMM for small M Test WebGPU Backend #1390: Pull request #20154 synchronize by digantdesai
9s gh/digantdesai/59/head
gh/digantdesai/59/head
9s
Update tokenizers submodule to the linear-time HF encode path Test WebGPU Backend #1389: Pull request #20349 opened by digantdesai
43m 25s gh/digantdesai/68/head
gh/digantdesai/68/head
43m 25s
Add chunked prefill to the eagle3 runner for prompts past the ring-cache cap Test WebGPU Backend #1388: Pull request #20347 opened by digantdesai
43m 13s gh/digantdesai/66/head
gh/digantdesai/66/head
43m 13s
Add --prompt_file flag to the eagle3 runner Test WebGPU Backend #1387: Pull request #20348 opened by digantdesai
42m 41s gh/digantdesai/67/head
gh/digantdesai/67/head
42m 41s
Make EAGLE-3 chain length K a runtime parameter Test WebGPU Backend #1386: Pull request #20345 opened by digantdesai
41m 28s gh/digantdesai/64/head
gh/digantdesai/64/head
41m 28s
Fix sliding-window chunked prefill in the gemma4-31B runner Test WebGPU Backend #1385: Pull request #20346 opened by digantdesai
40m 33s gh/digantdesai/65/head
gh/digantdesai/65/head
40m 33s
Add split-K mid-M SDPA Triton kernel for EAGLE-3 target_verify at long context Test WebGPU Backend #1384: Pull request #20344 opened by digantdesai
40m 42s gh/digantdesai/63/head
gh/digantdesai/63/head
40m 42s
[ET-VK] Add dynamic-shape resize to q8ta ops Test WebGPU Backend #1383: Pull request #20312 synchronize by SS-JIA
1h 10m 33s gh/SS-JIA/558/head
gh/SS-JIA/558/head
1h 10m 33s
[ET-VK][conv2d_dw] Fix test requesting nonexistent sned b1x1 shader Test WebGPU Backend #1382: Pull request #20343 opened by SS-JIA
1h 6m 34s gh/SS-JIA/559/head
gh/SS-JIA/559/head
1h 6m 34s
Update executorch-ubuntu-26.04-gcc15 docker image to ubuntu-26.04-gcc… Test WebGPU Backend #1381: Commit 315a0d8 pushed by rascani
56m 1s main
main
56m 1s
[MLX] Reduce physical footprint memory in MLX Test WebGPU Backend #1380: Pull request #20342 opened by metascroy
59m 24s reduce-mem-slot-recycling-fix
reduce-mem-slot-recycling-fix
59m 24s
[MLX] Reduce physical footprint memory in RingBufferKVCache for chunked prefill Test WebGPU Backend #1379: Pull request #20341 opened by metascroy
1h 1m 43s reduce-mem-sliding-kv
reduce-mem-sliding-kv
1h 1m 43s
ProTip! You can narrow down the results and go further in time using created:<2026-06-17 or the other filters available.