Skip to content

Actions: pytorch/executorch

Actions

cuda-perf

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
cuda-perf.yml
520 workflow runs
520 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
cuda-perf cuda-perf #523: Scheduled
1h 13m 21s main
main
1h 13m 21s
Add Qwen 3.6 MoE model and switch CI to Qwen3.6-35B-A3B-HQQ-INT4 cuda-perf #522: Pull request #18978 synchronize by mergennachin
1h 8m 40s qwen3_6
qwen3_6
1h 8m 40s
Add Qwen 3.6 MoE model and switch CI to Qwen3.6-35B-A3B-HQQ-INT4 cuda-perf #521: Pull request #18978 synchronize by mergennachin
43m 15s qwen3_6
qwen3_6
43m 15s
Add Qwen 3.6 MoE model and switch CI to Qwen3.6-35B-A3B-HQQ-INT4 cuda-perf #520: Pull request #18978 synchronize by mergennachin
44m 59s qwen3_6
qwen3_6
44m 59s
Add Qwen 3.6 MoE model and switch CI to Qwen3.6-35B-A3B-HQQ-INT4 cuda-perf #519: Pull request #18978 opened by mergennachin
53m 51s qwen3_6
qwen3_6
53m 51s
cuda-perf cuda-perf #518: Scheduled
1h 8m 6s main
main
1h 8m 6s
Add CUDA graph capture/replay for qwen 3.5 moe decode method cuda-perf #517: Pull request #18809 synchronize by Gasoonjia
57m 0s cuda-graph
cuda-graph
57m 0s
Revert "Add Qwen 3.6 MoE model and switch CI to Qwen3.6-35B-A3B-HQQ-I… cuda-perf #516: Commit f4019c3 pushed by mergennachin
54m 29s main
main
54m 29s
Revert "Add Qwen 3.6 MoE model and switch CI to Qwen3.6-35B-A3B-HQQ-INT4" cuda-perf #515: Pull request #18965 opened by mergennachin
1h 12m 40s revert-18955-qwen3_6
revert-18955-qwen3_6
1h 12m 40s
Add Qwen 3.6 MoE model and switch CI to Qwen3.6-35B-A3B-HQQ-INT4 (#18… cuda-perf #514: Commit 7fdd306 pushed by mergennachin
1h 0m 34s main
main
1h 0m 34s
Add CUDA graph capture/replay for qwen 3.5 moe decode method cuda-perf #513: Pull request #18809 synchronize by Gasoonjia
1h 3m 14s cuda-graph
cuda-graph
1h 3m 14s
Add Qwen 3.6 MoE model and switch CI to Qwen3.6-35B-A3B-HQQ-INT4 cuda-perf #512: Pull request #18955 opened by mergennachin
1h 4m 44s qwen3_6
qwen3_6
1h 4m 44s
Per weight constant cache cuda-perf #511: Pull request #18901 synchronize by Gasoonjia
58m 57s per-weight-constant-cache
per-weight-constant-cache
58m 57s
Add CUDA graph capture/replay for qwen 3.5 moe decode method cuda-perf #510: Pull request #18809 synchronize by Gasoonjia
29m 20s cuda-graph
cuda-graph
29m 20s
cuda-perf cuda-perf #509: Scheduled
1h 15m 39s main
main
1h 15m 39s
Per weight constant cache cuda-perf #508: Pull request #18901 synchronize by Gasoonjia
54m 27s per-weight-constant-cache
per-weight-constant-cache
54m 27s
Add CUDA graph capture/replay for qwen 3.5 moe decode method cuda-perf #507: Pull request #18809 synchronize by Gasoonjia
29m 4s cuda-graph
cuda-graph
29m 4s
Per weight constant cache cuda-perf #506: Pull request #18901 synchronize by Gasoonjia
57m 31s per-weight-constant-cache
per-weight-constant-cache
57m 31s
Per weight constant cache cuda-perf #505: Pull request #18901 synchronize by Gasoonjia
1h 6m 49s per-weight-constant-cache
per-weight-constant-cache
1h 6m 49s
Add Qwen 3.5 MoE to cuda-perf CI and add prefill throughput tracking cuda-perf #504: Pull request #18903 synchronize by Gasoonjia
2h 35m 9s cuda-perf-qwen35-moe
cuda-perf-qwen35-moe
2h 35m 9s
Add Qwen 3.5 MoE to cuda-perf CI and add prefill throughput tracking cuda-perf #503: Pull request #18903 synchronize by Gasoonjia
21m 11s cuda-perf-qwen35-moe
cuda-perf-qwen35-moe
21m 11s
Add CUDA graph capture/replay for qwen 3.5 moe decode method cuda-perf #502: Pull request #18809 synchronize by Gasoonjia
56m 37s cuda-graph
cuda-graph
56m 37s
Add CUDA graph capture/replay for qwen 3.5 moe decode method cuda-perf #501: Pull request #18809 synchronize by Gasoonjia
54m 48s cuda-graph
cuda-graph
54m 48s
Add Qwen 3.5 MoE to cuda-perf CI and add prefill throughput tracking cuda-perf #500: Pull request #18903 synchronize by Gasoonjia
2h 14m 21s cuda-perf-qwen35-moe
cuda-perf-qwen35-moe
2h 14m 21s
Add Qwen 3.5 MoE to cuda-perf CI and add prefill throughput tracking cuda-perf #499: Pull request #18903 synchronize by Gasoonjia
8m 47s cuda-perf-qwen35-moe
cuda-perf-qwen35-moe
8m 47s