Skip to content

Actions: pytorch/executorch

Actions

MLX

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
mlx.yml
910 workflow runs
910 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 CUDA graph capture/replay for qwen 3.5 moe decode method MLX #235: Pull request #18809 synchronize by Gasoonjia
53m 44s cuda-graph
cuda-graph
53m 44s
Add fused GatedDeltaNet decode Triton kernel MLX #234: Pull request #18865 synchronize by Gasoonjia
51m 43s fused-deltanet-decode
fused-deltanet-decode
51m 43s
fix format str (#18916) MLX #233: Commit 2f339f0 pushed by digantdesai
35m 26s main
main
35m 26s
Per weight constant cache MLX #232: Pull request #18901 synchronize by Gasoonjia
36m 7s per-weight-constant-cache
per-weight-constant-cache
36m 7s
Per weight constant cache MLX #231: Pull request #18901 synchronize by Gasoonjia
32m 41s per-weight-constant-cache
per-weight-constant-cache
32m 41s
[ET-VK] Fix use-after-free in PrepackNode when TensorRefs are shared MLX #230: Commit f84a17c pushed by SS-JIA
37m 47s main
main
37m 47s
Add Qwen 3.5 MoE to cuda-perf CI and add prefill throughput tracking MLX #229: Pull request #18903 synchronize by Gasoonjia
39m 26s cuda-perf-qwen35-moe
cuda-perf-qwen35-moe
39m 26s
Add Qwen 3.5 MoE to cuda-perf CI and add prefill throughput tracking MLX #228: Pull request #18903 synchronize by Gasoonjia
33m 43s cuda-perf-qwen35-moe
cuda-perf-qwen35-moe
33m 43s
Add Pico2 build validation CI (trunk only) (#18900) MLX #227: Commit 28c56fe pushed by psiddh
46m 2s main
main
46m 2s
Add CUDA graph capture/replay for qwen 3.5 moe decode method MLX #226: Pull request #18809 synchronize by Gasoonjia
37m 39s cuda-graph
cuda-graph
37m 39s
Add CUDA graph capture/replay for qwen 3.5 moe decode method MLX #225: Pull request #18809 synchronize by Gasoonjia
38m 43s cuda-graph
cuda-graph
38m 43s
Qualcomm AI Engine Direct - Adding QNN backend support for the remain… MLX #224: Commit f35fbb5 pushed by abhinaykukkadapu
51m 29s main
main
51m 29s
Add Qwen 3.5 MoE to cuda-perf CI and add prefill throughput tracking MLX #223: Pull request #18903 synchronize by Gasoonjia
1h 28m 31s cuda-perf-qwen35-moe
cuda-perf-qwen35-moe
1h 28m 31s
Add Qwen 3.5 MoE to cuda-perf CI and add prefill throughput tracking MLX #222: Pull request #18903 synchronize by Gasoonjia
1h 16m 16s cuda-perf-qwen35-moe
cuda-perf-qwen35-moe
1h 16m 16s
[Parakeet] Add Vulkan backend documentation and fix CMake build MLX #221: Commit eaef2ed pushed by SS-JIA
43m 33s main
main
43m 33s
[Parakeet] Add Vulkan backend documentation and fix CMake build MLX #220: Pull request #18734 synchronize by SS-JIA
38m 11s pr18734
pr18734
38m 11s
[Parakeet] Add Vulkan backend documentation and fix CMake build MLX #219: Pull request #18734 synchronize by SS-JIA
39m 11s pr18734
pr18734
39m 11s
[Parakeet] Add Vulkan backend documentation and fix CMake build MLX #218: Pull request #18734 synchronize by SS-JIA
37m 58s pr18734
pr18734
37m 58s
[Parakeet] Add Vulkan backend documentation and fix CMake build MLX #217: Pull request #18734 synchronize by SS-JIA
51m 50s pr18734
pr18734
51m 50s
Fix GRU w8a32 operator (#17226) MLX #216: Commit f1209c5 pushed by meta-codesync Bot
41m 0s main
main
41m 0s
Fix Conv1d w8a32 operator (#16607) MLX #215: Commit a8ca3f0 pushed by meta-codesync Bot
1h 1m 20s main
main
1h 1m 20s
Disable VGF tests (#18899) MLX #214: Commit 75fe8e9 pushed by meta-codesync Bot
35m 29s main
main
35m 29s
[ET-VK] Lower reduce_peak_memory threshold from 500 MB to 10 MB MLX #213: Commit 89e6416 pushed by SS-JIA
39m 12s main
main
39m 12s
MoE prefill bf16 perf improvement for qwen-3.5-35B-A3B (#18829) MLX #212: Commit 8d1ff1a pushed by digantdesai
32m 14s main
main
32m 14s
Arm backend: Eliminate dead shape ops (#18902) MLX #211: Commit b5cf3c3 pushed by oscarandersson8218
35m 37s main
main
35m 37s
ProTip! You can narrow down the results and go further in time using created:<2026-04-15 or the other filters available.