Skip to content

Actions: deepseek-ai/DeepEP

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
38 workflow runs
38 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Enable CUDA Graph for internode dispatch (#438) Code Format Check #38: Commit 92fe2de pushed by sphish
29s main
main
29s
Support CUDA Graph for internode dispatch normal kernel Code Format Check #37: Pull request #438 synchronize by yifeizhang-c
25s yifeizhang-c:enable-internode-cuda-graph
yifeizhang-c:enable-internode-cuda-graph
25s
Support CUDA Graph for internode dispatch normal kernel Code Format Check #36: Pull request #438 synchronize by yifeizhang-c
25s yifeizhang-c:enable-internode-cuda-graph
yifeizhang-c:enable-internode-cuda-graph
25s
Support cuMem API in cross process shared memory management (#217) Code Format Check #35: Commit 9249c25 pushed by LyricZhao
23s main
main
23s
Add barrier before nvshmem_team_split_strided to ensure proper initialization Code Format Check #34: Pull request #478 opened by pierDipi
30s pierDipi:barrier-low-latency
pierDipi:barrier-low-latency
30s
Add imbalance factor in test_low_latency Code Format Check #33: Pull request #393 synchronize by JianboDong
Action required JianboDong:add_imblance
JianboDong:add_imblance
Action required
Allow NVSHMEM PE to NIC mapping to be initialized by DeepEP rank Code Format Check #32: Pull request #466 synchronize by smarterclayton
Action required smarterclayton:nic_pe_alignment
smarterclayton:nic_pe_alignment
Action required
Compatible with older versions of Python and Torch (#469) Code Format Check #31: Commit bfded34 pushed by LyricZhao
22s main
main
22s
Compatible with older versions of Python and Torch Code Format Check #30: Pull request #469 synchronize by sphish
24s compat-old-torch
compat-old-torch
24s
Compatible with older versions of Python and Torch Code Format Check #29: Pull request #469 opened by sphish
23s compat-old-torch
compat-old-torch
23s
Support CUDA Graph for internode dispatch normal kernel Code Format Check #28: Pull request #438 synchronize by yifeizhang-c
27s yifeizhang-c:enable-internode-cuda-graph
yifeizhang-c:enable-internode-cuda-graph
27s
Support CUDA Graph for internode dispatch normal kernel Code Format Check #27: Pull request #438 synchronize by yifeizhang-c
28s yifeizhang-c:enable-internode-cuda-graph
yifeizhang-c:enable-internode-cuda-graph
28s
Support cuMem API in cross process shared memory management Code Format Check #26: Pull request #217 synchronize by fzyzcjy
20s fzyzcjy:feat/cu_mem_api
fzyzcjy:feat/cu_mem_api
20s
Support cuMem API in cross process shared memory management Code Format Check #25: Pull request #217 synchronize by fzyzcjy
22s fzyzcjy:feat/cu_mem_api
fzyzcjy:feat/cu_mem_api
22s
Support cuMem API in cross process shared memory management Code Format Check #24: Pull request #217 synchronize by fzyzcjy
22s fzyzcjy:feat/cu_mem_api
fzyzcjy:feat/cu_mem_api
22s
Support CUDA Graph for internode dispatch normal kernel Code Format Check #23: Pull request #438 synchronize by yifeizhang-c
20s yifeizhang-c:enable-internode-cuda-graph
yifeizhang-c:enable-internode-cuda-graph
20s
Support CUDA Graph for internode dispatch normal kernel Code Format Check #22: Pull request #438 synchronize by yifeizhang-c
31s yifeizhang-c:enable-internode-cuda-graph
yifeizhang-c:enable-internode-cuda-graph
31s
Allow NVSHMEM PE to NIC mapping to be initialized by DeepEP rank Code Format Check #21: Pull request #466 synchronize by smarterclayton
Action required smarterclayton:nic_pe_alignment
smarterclayton:nic_pe_alignment
Action required
Allow NVSHMEM PE to NIC mapping to be initialized by DeepEP rank Code Format Check #20: Pull request #466 opened by smarterclayton
Action required smarterclayton:nic_pe_alignment
smarterclayton:nic_pe_alignment
Action required
Update README (#463) Code Format Check #19: Commit 2be0d4f pushed by LyricZhao
32s main
main
32s
Add Tencent's zero-copy branch into README Code Format Check #18: Pull request #463 opened by LyricZhao
29s tencent-zcopy
tencent-zcopy
29s
fix benchmark error when slow Code Format Check #17: Pull request #460 opened by fishautumn
Action required fishautumn:feat/units
fishautumn:feat/units
Action required
support hidden-dim 3072 (#458) Code Format Check #16: Commit 73b6ea4 pushed by sphish
28s main
main
28s
Support hidden-dim 3072 for gpt-oss + low_latency kernels Code Format Check #15: Pull request #458 opened by varun-sundar-rabindranath
28s varun-sundar-rabindranath:varun/gptoss-padding
varun-sundar-rabindranath:varun/gptoss-padding
28s
feat: remove recordStream of normal mode Code Format Check #14: Pull request #456 synchronize by junjzhang
Action required junjzhang:feat/remove_record_stream
junjzhang:feat/remove_record_stream
Action required