Skip to content

Actions: pytorch/executorch

Actions

MLX

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
mlx.yml
516 workflow runs
516 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
Back out "Re-apply D101260086: Android unified error reporting" (#19137) MLX #518: Commit bf64fa1 pushed by meta-codesync Bot
35m 51s main
main
35m 51s
Re-apply D101260086: Android unified error reporting MLX #517: Commit 563be2f pushed by meta-codesync Bot
1h 8m 38s main
main
1h 8m 38s
Add top-k sampling support to llm Sampler (#19122) MLX #516: Commit 2d9bbc1 pushed by meta-codesync Bot
33m 53s main
main
33m 53s
Back out D102011505 and D101260086 (#19134) MLX #515: Commit bdf1bf4 pushed by meta-codesync Bot
35m 11s main
main
35m 11s
Revert Android PRs #19099, #19124, #19092, #19028 (#19133) MLX #514: Commit 6b175ff pushed by JacobSzwejbka
41m 27s main
main
41m 27s
Add C++ unit tests for cadence::quantized_conv2d_nhwc + add depthwise… MLX #513: Commit 5252704 pushed by meta-codesync Bot
37m 9s main
main
37m 9s
Merge back to back slices on the same dim MLX #512: Commit c1d482e pushed by meta-codesync Bot
35m 41s main
main
35m 41s
Android: consistent error types across all modules (#19099) MLX #511: Commit 7e2ff8a pushed by psiddh
34m 47s main
main
34m 47s
Android: Module implements Closeable (#19124) MLX #510: Commit b8f04aa pushed by psiddh
35m 43s main
main
35m 43s
Use safe numel() in ET (retake) (#19130) MLX #509: Commit e1cd352 pushed by lucylq
41m 40s main
main
41m 40s
Add safe_numel() MLX #508: Commit 222711e pushed by meta-codesync Bot
33m 14s main
main
33m 14s
MLX delegate: add integer support for aten.bitwise_not (#19053) MLX #507: Commit 0a43e2f pushed by metascroy
37m 30s main
main
37m 30s
Allow chunked prefill when num_prompt_tokens > max_seq_len MLX #506: Commit 2330652 pushed by meta-codesync Bot
35m 18s main
main
35m 18s
portable: accumulate in fp32 for Half/BFloat16 in grid_sampler_2d bil… MLX #505: Commit 60ffe19 pushed by GregoryComer
37m 41s main
main
37m 41s
MLX delegate: add integer support for aten.bitwise_not MLX #504: Pull request #19053 synchronize by metascroy
1h 3m 58s AlessandroVacca:mlx-bitwise-not-int
AlessandroVacca:mlx-bitwise-not-int
1h 3m 58s
Numel and Nbytes Validation (#19121) (#19121) MLX #503: Commit 56da964 pushed by JacobSzwejbka
39m 24s main
main
39m 24s
Qualcomm AI Engine Direct - Fix depth_anything (#19058) MLX #502: Commit cb94506 pushed by abhinaykukkadapu
39m 48s main
main
39m 48s
Add fused GatedDeltaNet decode Triton kernel MLX #501: Pull request #18865 synchronize by Gasoonjia
35m 50s fused-deltanet-decode
fused-deltanet-decode
35m 50s
Add GPU-side Gumbel-max sampling for CUDA graph compatibility MLX #500: Pull request #18844 synchronize by Gasoonjia
1h 3m 27s cuda-graph-sampling
cuda-graph-sampling
1h 3m 27s
Use EValue::tryTo* (#19040) MLX #499: Commit de8ce55 pushed by lucylq
35m 32s main
main
35m 32s
Fix overflows in et (#19057) MLX #498: Commit ec5e8e4 pushed by lucylq
40m 15s main
main
40m 15s
Add recurrent gated delta rule custom op for Qwen3.5 attention (#18088) MLX #497: Commit 476a7ef pushed by digantdesai
35m 33s main
main
35m 33s
Remove un-used copy from building dockers (#18868) MLX #496: Commit 98a1d66 pushed by Erik-Lundell
1h 7m 56s main
main
1h 7m 56s
Arm backend: Add util for symbolic range eval (#19108) MLX #495: Commit c5c5b3a pushed by oscarandersson8218
53m 30s main
main
53m 30s
Arm backend: Disable fusing of TOSA ops (#19066) MLX #494: Commit b6a47aa pushed by oscarandersson8218
49m 56s main
main
49m 56s