Skip to content

Actions: pytorch/executorch

Actions

cuda-perf

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
cuda-perf.yml
573 workflow runs
573 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
Merge back to back slices on the same dim cuda-perf #576: Commit c1d482e pushed by meta-codesync Bot
1h 58m 25s main
main
1h 58m 25s
Android: consistent error types across all modules (#19099) cuda-perf #575: Commit 7e2ff8a pushed by psiddh
2h 9m 58s main
main
2h 9m 58s
Android: Module implements Closeable (#19124) cuda-perf #574: Commit b8f04aa pushed by psiddh
2h 27m 50s main
main
2h 27m 50s
Use safe numel() in ET (retake) (#19130) cuda-perf #573: Commit e1cd352 pushed by lucylq
2h 41m 45s main
main
2h 41m 45s
Add safe_numel() cuda-perf #572: Commit 222711e pushed by meta-codesync Bot
2h 45m 34s main
main
2h 45m 34s
MLX delegate: add integer support for aten.bitwise_not (#19053) cuda-perf #571: Commit 0a43e2f pushed by metascroy
2h 11m 32s main
main
2h 11m 32s
Allow chunked prefill when num_prompt_tokens > max_seq_len cuda-perf #570: Commit 2330652 pushed by meta-codesync Bot
2h 51m 56s main
main
2h 51m 56s
portable: accumulate in fp32 for Half/BFloat16 in grid_sampler_2d bil… cuda-perf #569: Commit 60ffe19 pushed by GregoryComer
2h 22m 36s main
main
2h 22m 36s
Numel and Nbytes Validation (#19121) (#19121) cuda-perf #568: Commit 56da964 pushed by JacobSzwejbka
2h 1m 39s main
main
2h 1m 39s
Qualcomm AI Engine Direct - Fix depth_anything (#19058) cuda-perf #567: Commit cb94506 pushed by abhinaykukkadapu
2h 39m 52s main
main
2h 39m 52s
Use EValue::tryTo* (#19040) cuda-perf #566: Commit de8ce55 pushed by lucylq
2h 54m 44s main
main
2h 54m 44s
Fix overflows in et (#19057) cuda-perf #565: Commit ec5e8e4 pushed by lucylq
2h 31m 32s main
main
2h 31m 32s
Add recurrent gated delta rule custom op for Qwen3.5 attention (#18088) cuda-perf #564: Commit 476a7ef pushed by digantdesai
2h 25m 9s main
main
2h 25m 9s
Remove un-used copy from building dockers (#18868) cuda-perf #563: Commit 98a1d66 pushed by Erik-Lundell
2h 17m 6s main
main
2h 17m 6s
Arm backend: Add util for symbolic range eval (#19108) cuda-perf #562: Commit c5c5b3a pushed by oscarandersson8218
3h 36m 59s main
main
3h 36m 59s
Arm backend: Disable fusing of TOSA ops (#19066) cuda-perf #561: Commit b6a47aa pushed by oscarandersson8218
2h 55m 27s main
main
2h 55m 27s
skip cuda operations when running qwen 3.5 moe on other backend (#19095) cuda-perf #560: Commit b6cec38 pushed by Gasoonjia
2h 36m 22s main
main
2h 36m 22s
Widen resolve_max_new_tokens parameters to int64_t and rename for cla… cuda-perf #559: Commit eef7921 pushed by meta-codesync Bot
2h 5m 50s main
main
2h 5m 50s
Add tryTo evalue accessors (#19039) cuda-perf #558: Commit c3f3d12 pushed by lucylq
3h 58m 29s main
main
3h 58m 29s
Fix smollm2 alias to point at SmolLM2-135M (v2) instead of SmolLM-135… cuda-perf #557: Commit 75b31bb pushed by kirklandsign
3h 58m 50s main
main
3h 58m 50s
Validate XNNPACK tensor flags are valid (#19102) cuda-perf #556: Commit edb8c98 pushed by lucylq
3h 47m 53s main
main
3h 47m 53s
Add Half (float16) support to slim ScalarType enum (#18959) (#18959) cuda-perf #555: Commit 4a69750 pushed by kirklandsign
4h 1m 51s main
main
4h 1m 51s
Android: improve error diagnostics for LlmModule and exceptions (#19092) cuda-perf #554: Commit f9f29e7 pushed by psiddh
4h 15m 15s main
main
4h 15m 15s
Add add-relu fusion in the quantizer cuda-perf #553: Commit 7b5dcc1 pushed by meta-codesync Bot
3h 26m 21s main
main
3h 26m 21s
Extract shared multifunction PTE utilities to utils.py (#19035) cuda-perf #552: Commit 6d23e41 pushed by meta-codesync Bot
3h 0m 39s main
main
3h 0m 39s