Skip to content

Actions: flagos-ai/PyTorch-Plugin-FL

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
190 workflow runs
190 workflow runs

Filter by Workflow

⌥ + click/return to exclude

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
feat(bpu): D-Robotics RDK BPU backend with on-board compilation CI #184: Pull request #69 synchronize by lvyufeng
23m 15s lvyufeng:feat/s600-bpu-backend
lvyufeng:feat/s600-bpu-backend
23m 15s
feat(wheel): self-contained wheels that run on a stock torch+cpu CI #183: Pull request #68 opened by lvyufeng
54m 25s metax-2.10
metax-2.10
54m 25s
feat(gcu): run FlagGems Triton kernels on Enflame GCU (#65) CI #182: Commit 3c6529a pushed by zhaoyinglia
1h 29m 58s main
main
1h 29m 58s
feat(ascend): FlagGems Triton path on real 910 — 201 measured-correct… CI #181: Commit 09a70bf pushed by zhaoyinglia
1h 59m 19s main
main
1h 59m 19s
feat(gcu): run FlagGems Triton kernels on Enflame GCU CI #180: Pull request #65 opened by lvyufeng
1h 3m 32s lvyufeng:feat/flaggems-on-gcu
lvyufeng:feat/flaggems-on-gcu
1h 3m 32s
feat(musa): 13 reduction/normalization mudnn kernels + unbreak MUSA build CI #179: Pull request #63 synchronize by lvyufeng
2h 23m 55s feat/musa-p2-reductions
feat/musa-p2-reductions
2h 23m 55s
feat(ascend): FlagGems Triton path on real 910 — 201 measured-correct routes CI #178: Pull request #64 opened by lvyufeng
16h 40m 36s lvyufeng:npu-2.10
lvyufeng:npu-2.10
16h 40m 36s
feat(tileops): integrate TileOPs as an optional operator library (60 codegen'd routes) CI #177: Pull request #44 synchronize by lvyufeng
4h 22m 34s lvyufeng:feat/tileops-optional-backend
lvyufeng:feat/tileops-optional-backend
4h 22m 34s
docs(tests): replace machine-local HF_HOME paths with a placeholder (… CI #176: Commit a02f62b pushed by zhaoyinglia
7h 57m 43s main
main
7h 57m 43s
fix(comm): run collectives on their operand's device (#57) CI #175: Commit 68e3c2e pushed by zhaoyinglia
5m 51s main
main
5m 51s
feat(metax): FSDP2 full feature parity + Qwen3 training match (#62) CI #174: Commit 43c8985 pushed by zhaoyinglia
1h 30m 19s main
main
1h 30m 19s
feat(musa): 13 reduction/normalization mudnn kernels + unbreak MUSA build CI #173: Pull request #63 opened by lvyufeng
5h 27m 36s feat/musa-p2-reductions
feat/musa-p2-reductions
5h 27m 36s
feat(musa): 26 training-hot-path mudnn kernels + fix two Reduce bugs … CI #172: Commit e33f955 pushed by zhaoyinglia
41m 0s main
main
41m 0s
fix(comm): run collectives on their operand's device CI #171: Pull request #57 synchronize by lvyufeng
3h 47m 33s lvyufeng:fix-comm-device-index
lvyufeng:fix-comm-device-index
3h 47m 33s
fix(comm): run collectives on their operand's device CI #170: Pull request #57 synchronize by lvyufeng
12m 0s lvyufeng:fix-comm-device-index
lvyufeng:fix-comm-device-index
12m 0s
feat(metax): FSDP2 full feature parity + Qwen3 training match CI #169: Pull request #62 opened by lvyufeng
5h 28m 12s feat/metax-fsdp2
feat/metax-fsdp2
5h 28m 12s
feat(musa): 26 training-hot-path mudnn kernels CI #168: Pull request #60 opened by lvyufeng
2h 42m 35s feat/musa-p1-hot-path
feat/musa-p1-hot-path
2h 42m 35s
fix(comm): run collectives on their operand's device CI #167: Pull request #57 synchronize by lvyufeng
2h 25m 49s lvyufeng:fix-comm-device-index
lvyufeng:fix-comm-device-index
2h 25m 49s
fix(compile): make autotuning and parallel kernel compilation work on flagos CI #166: Pull request #59 synchronize by lvyufeng
3h 56m 21s lvyufeng:fix/compile-autotune-and-workers
lvyufeng:fix/compile-autotune-and-workers
3h 56m 21s
docs(tests): replace machine-local HF_HOME paths with a placeholder CI #165: Pull request #58 opened by lvyufeng
2h 7m 6s lvyufeng:fix/profiler-doc-abs-path
lvyufeng:fix/profiler-doc-abs-path
2h 7m 6s
fix(comm): run collectives on their operand's device CI #164: Pull request #57 opened by lvyufeng
18h 39m 3s lvyufeng:fix-comm-device-index
lvyufeng:fix-comm-device-index
18h 39m 3s
feat(profiler): torch-cuda parity for PrivateUse1 — flows, device tim… CI #163: Commit 11457c9 pushed by zhaoyinglia
2h 42m 14s main
main
2h 42m 14s
perf+fix: _to_copy/boxing hot-path opts + unfold view op & group-norm… CI #162: Commit d530681 pushed by zhaoyinglia
55m 18s main
main
55m 18s
feat(metax): enable the FlagGems C++ (kFlagOs) dispatch path (#56) CI #161: Commit 8695f5a pushed by zhaoyinglia
1h 15m 7s main
main
1h 15m 7s
perf+fix: _to_copy/boxing hot-path opts + unfold view op & group-norm layout fixes CI #160: Pull request #51 synchronize by shallitbeso
58m 39s shallitbeso:perf/dispatch-copy-boxing
shallitbeso:perf/dispatch-copy-boxing
58m 39s