Skip to content

Actions: pytorch/executorch

Actions

MLX

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
mlx.yml
2,500+ workflow runs
2,500+ 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
Arm backend: Generate runner-specific operator libs (#21615) MLX #2569: Commit e5f6b69 pushed by Erik-Lundell
37s main
main
37s
Arm backend: Replace model converter check with preflight VGF tool (#… MLX #2568: Commit e151214 pushed by wwwind
37s main
main
37s
[Qualcomm] Scope ExpandBroadcastTensorShape rewrite to the broadcast … MLX #2567: Commit ae227c2 pushed by psiddh
43s main
main
43s
Arm backend: Various build fixes (#21641) MLX #2566: Commit 812c7f0 pushed by Erik-Lundell
59m 52s main
main
59m 52s
Arm backend: Keep DeepSeek FP8 on xlarge path (#21612) MLX #2565: Commit 70594d3 pushed by bdemirb
39s main
main
39s
Fix IndexError when a squeeze view is reached upstream (#21620) MLX #2564: Commit 28a7fac pushed by meta-codesync Bot
37s main
main
37s
[MLX] Fix HF sliding-window KV cache sizing and enable ring buffers MLX #2563: Pull request #21635 opened by kiymetakdemir
34m 37s kiymetakdemir:fix-gemma3-sliding-window
kiymetakdemir:fix-gemma3-sliding-window
34m 37s
Add large CUDA DFlash target prefill and small-query SDPA (#21628) MLX #2562: Commit 82fcf09 pushed by meta-codesync Bot
34s main
main
34s
[RELEASE ONLY CHANGES] Require PyTorch for full wheel (#21632) MLX #2561: Commit 3dd7ccd pushed by JacobSzwejbka
36s release/1.4
release/1.4
36s
[RELEASE ONLY CHANGES] Require PyTorch for full wheel MLX #2560: Commit ac877f5 pushed by JacobSzwejbka
5m 41s release/1.4-full-wheel-torch-dependency
release/1.4-full-wheel-torch-dependency
5m 41s
Stop gathering yolo26 for the portable CI matrix (#21618) MLX #2559: Commit 99924e4 pushed by shoumikhin
48m 10s main
main
48m 10s
Keep DS-CNN example inputs portable-safe (#21619) MLX #2558: Commit 74f1ce8 pushed by shoumikhin
1h 53m 25s main
main
1h 53m 25s
CoreML-safe RMSNormWithInputScale for fp16 export MLX #2557: Commit bcceab1 pushed by meta-codesync Bot
3h 29m 5s main
main
3h 29m 5s
Fix concurrency key so ciflow tag re-pushes cancel prior runs MLX #2556: Pull request #21622 opened by huydhn
4h 40m 57s fix-ciflow-tag-concurrency-cancellation
fix-ciflow-tag-concurrency-cancellation
4h 40m 57s
Run coreml_static_int8 macOS tests on the nightly schedule only (#21609) MLX #2555: Commit 85ade9f pushed by shoumikhin
25m 39s main
main
25m 39s
Arm backend: Enable using minimal c++ runtime. (#21587) MLX #2554: Commit b21661f pushed by Erik-Lundell
1h 9m 36s main
main
1h 9m 36s
Arm backend: preserve Q/DQ for partially quant ops (#21589) MLX #2553: Commit a931a27 pushed by zingo
41s main
main
41s
[RFC] DFlash MLX #2552: Pull request #20840 synchronize by cthotti
Action required cthotti:dflash-qwen3-4b
cthotti:dflash-qwen3-4b
Action required
[RFC] DFlash MLX #2551: Pull request #20840 synchronize by cthotti
Action required cthotti:dflash-qwen3-4b
cthotti:dflash-qwen3-4b
Action required
Codegen: Support empty selected operator sets (#21548) MLX #2550: Commit 2afdd93 pushed by Erik-Lundell
37s main
main
37s
NXP backend: Add support for softmax with the new Neutron flow. (#1… MLX #2549: Commit b61ebdd pushed by irtrukhina
40s main
main
40s
Unconditional in-kernel causal attention mask (long-context prefill speedup) (#20895) MLX #2548: Pull request #20895 synchronize by meta-codesync Bot
53m 21s export-D111546133
export-D111546133
53m 21s
Stop using deprecated executorch PassManager (#21576) MLX #2547: Commit 8702327 pushed by meta-codesync Bot
56m 11s main
main
56m 11s
Dedupe getitem nodes in ExportedProgramPassManager (#21590) MLX #2546: Commit b744a61 pushed by meta-codesync Bot
45s main
main
45s
Test permutation-sink view that splits the non-unit dim MLX #2545: Commit 8422afd pushed by meta-codesync Bot
41s main
main
41s