Skip to content

Actions: pytorch/executorch

Actions

MLX

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
mlx.yml
2,345 workflow runs
2,345 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
Record LLM model execution latency (#21040) MLX #2294: Commit 776ba0c pushed by JakeStevens
42s main
main
42s
Use dynamic CPU count for cmake --build -j in docs and test scripts (… MLX #2293: Commit 7f86746 pushed by JakeStevens
37s main
main
37s
Cortex-M backend: lower quantized GELU via the activation LUT (#21025) MLX #2292: Commit 7628aa6 pushed by rascani
45m 53s main
main
45m 53s
Arm backend: Document RIFE accuracy handover (#21059) MLX #2291: Commit 53180df pushed by bdemirb
41s main
main
41s
Arm backend: Support SNORM RGBA runtime format (#20964) MLX #2290: Commit 4f3f8ed pushed by robell
39s main
main
39s
[ET-VK][sdpa] Vendor-adaptive head_dim output-tiling (TILE_N4=2) in G… MLX #2288: Commit 266e0dc pushed by SS-JIA
36s main
main
36s
[ExecuTorch][llm] Wire --use_moe_quantized_op through export pipeline and LLM config MLX #2287: Pull request #21126 opened by digantdesai
38m 15s gh/digantdesai/76/head
gh/digantdesai/76/head
38m 15s
Fuse consecutive clamps into a single clamp (#21013) MLX #2286: Commit b22a902 pushed by meta-codesync Bot
44s main
main
44s
Fix viable strict updating job (#21111) MLX #2285: Commit 181ded4 pushed by metascroy
32s main
main
32s
Flakey CI fix (#21110) MLX #2284: Commit f6eaba2 pushed by metascroy
37m 50s main
main
37m 50s
Qualcomm AI Engine Direct - improve llama3.2 3B TPS (#20903) MLX #2283: Commit fcc1020 pushed by psiddh
35s main
main
35s
up MLX #2282: Commit 8ab734a pushed by pytorch-bot Bot
48m 32s ciflow/mlx/21110
ciflow/mlx/21110
48m 32s
Flakey CI fix MLX #2281: Pull request #21110 opened by metascroy
47m 37s flakey-ci-fix
flakey-ci-fix
47m 37s
[ExecuTorch][WebGPU] Fuse attention QKV projections in prefill (#21107) MLX #2280: Commit 5828778 pushed by JCNTH
42s main
main
42s
Preserve BatchNorm running state in composable quantizer MLX #2279: Commit 6bd0355 pushed by denizkilinc
42s main
main
42s
Revert "Arm backend: Rerun duplicate-user fusion after TOSA lowering"… MLX #2278: Commit cd380e7 pushed by metascroy
42m 35s main
main
42m 35s
Make extension/cuda:caller_stream fbcode-only MLX #2277: Commit 37400d9 pushed by meta-codesync Bot
41s main
main
41s
gate fbcode-only TARGETS dirs with is_fbcode() and rename to BUCK MLX #2276: Commit b3ac2df pushed by meta-codesync Bot
44s main
main
44s
Add cooperative matrix dispatch for quantized linear MLX #2275: Commit 4a95dff pushed by meta-codesync Bot
36s main
main
36s
[ExecuTorch][WebGPU] Add fp32 matmul ops (mm, bmm, linear) to the Web… MLX #2274: Commit 666ba0b pushed by JCNTH
10m 25s main
main
10m 25s
Proper implementation for VulkanBackend::is_available() MLX #2273: Commit 8b14fc2 pushed by meta-codesync Bot
40s main
main
40s
Fix embedding lowering crash when QNN_SDK_ROOT is unset (#21051) MLX #2272: Commit 5ed466e pushed by psiddh
38s main
main
38s
Revert "Two-pass optimization for channelwise gated delta rule kernel… MLX #2271: Commit 0fc1f61 pushed by metascroy
48s main
main
48s
Refactor quantization into a shared extension/llm/export/quant packag… MLX #2270: Commit 110ac65 pushed by metascroy
38m 38s main
main
38m 38s
More flexible MoE support mobile focused (#21047) MLX #2269: Commit bfed808 pushed by meta-codesync Bot
3m 1s main
main
3m 1s