Skip to content

Actions: CodeLinaro/executorch

Actions

MLX

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
mlx.yml
128 workflow runs
128 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
Qualcomm AI Engine Direct - Fix a QNN crash on AMD (#22543) MLX #128: Commit 9036d84 pushed by zhaoxul-qti
 16s main
main
 16s
Arm backend: Delegate reflection padding on U55 MLX #127: Commit 843f77e pushed by qti-horodnic
 16s main
main
 16s
Arm backend: Add pre-decomposition partitioner pipeline (#22514) MLX #126: Commit c8d5189 pushed by quic-boyuc
 19s main
main
 19s
Enable use_sdpa_with_kv_cache in the qwen3_5 example config (#22390) MLX #125: Commit 5b829f9 pushed by qti-horodnic
 16s main
main
 16s
Qualcomm AI Engine Direct - [GenAI Pipeline] PR6: Compilation & infer… MLX #124: Commit c27baa8 pushed by qti-horodnic
 17s main
main
 17s
arm: extend CMSIS-Pack smoke test to Cortex-M (#22113) MLX #123: Commit 7f4b8ad pushed by qti-horodnic
 18s main
main
 18s
Skip redundant recompile after permute propagation retrace (#21696) MLX #122: Commit d0c2da3 pushed by qti-horodnic
 15s main
main
 15s
Fold DyT affine maps into following convolutions (#21953) (#21953) MLX #121: Commit a77eb9f pushed by qti-horodnic
 1m 12s main
main
 1m 12s
Arm backend: Remove RESIZE that don't change shape MLX #120: Commit 41d29de pushed by quic-boyuc
 16s main
main
 16s
Ship the CUDA Windows import library only in CUDA builds (#22054) MLX #119: Commit cff6f4d pushed by chenweng-quic
 16s main
main
 16s
Channels last: preserve metadata across op replacement (#21961) MLX #118: Commit ed7a23e pushed by qti-horodnic
 20m 45s main
main
 20m 45s
Qualcomm: give custom op args with no QNN mapping a real error (#21868) MLX #117: Commit b45e85d pushed by shewu-quic
 18s main
main
 18s
Support extra quantizable ops in advance quant above def chain (#21757) MLX #116: Commit a1eee71 pushed by qti-horodnic
 19s main
main
 19s
Fix XNNPACK rejecting avg_pool2d with default arguments (#21862) MLX #115: Commit 9ffbe36 pushed by qti-horodnic
 33s main
main
 33s
Reseed the Cortex-M test RNG before each test module is imported (#21… MLX #114: Commit 3253a28 pushed by qti-horodnic
 27s main
main
 27s
Share selective prim-ops helper (#21732) MLX #113: Commit 7857dfd pushed by chenweng-quic
 18s main
main
 18s
Qualcomm AI Engine Direct - Add QNN HadamardTransform op support (#21… MLX #112: Commit 43f89fb pushed by qti-horodnic
 31s main
main
 31s
Arm backend: Fix pad transpose handling around grid sampler (#21738) MLX #111: Commit c56e6bf pushed by qti-horodnic
 15s main
main
 15s
Qualcomm AI Engine Direct - Update quant recipe for PyTorch version b… MLX #110: Commit 5a77b43 pushed by shewu-quic
 23s main
main
 23s
Infer XNNPACK conv dimensionality from the weight rank (#21588) MLX #109: Commit 9cd0c12 pushed by qti-horodnic
 34s main
main
 34s
Optimize CUDA export host memory usage (#21617) MLX #108: Commit a439afd pushed by qti-horodnic
 34s main
main
 34s
Adjust how nax are excluded on < macOS 26.0 (#21554) MLX #107: Commit b4ffdbe pushed by qti-horodnic
 20s main
main
 20s
Add end-to-end sampling controls to the LLM server worker protocol (#… MLX #106: Commit 454f979 pushed by qti-horodnic
 18s main
main
 18s
[MLX] Add off-graph KV cache ring runtime (#21532) MLX #105: Commit 6bbb75a pushed by qti-horodnic
 38s main
main
 38s
Fix DS-CNN Cortex-M input memory format (#21485) MLX #104: Commit d632341 pushed by qti-horodnic
 19s main
main
 19s