Skip to content

Actions: deepseek-ai/DeepEP

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
128 workflow runs
128 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
Fix TypeError in cached ElasticBuffer dispatch with default num_sms Code Format Check #244: Pull request #685 synchronize by yurekami
21s yurekami:ghcontrib/elastic-cached-num-sms
yurekami:ghcontrib/elastic-cached-num-sms
21s
Fix TypeError in cached ElasticBuffer dispatch with default num_sms Code Format Check #243: Pull request #685 synchronize by yurekami
27s yurekami:ghcontrib/elastic-cached-num-sms
yurekami:ghcontrib/elastic-cached-num-sms
27s
docs: point Development section to develop.sh (fix stale ln -s for V2) Code Format Check #242: Pull request #741 opened by pengfaming
-1s pengfaming:fix/readme-development-ln-s
pengfaming:fix/readme-development-ln-s
-1s
Add unordered hybrid kernels for transports without ordered delivery (opt-in via EP_HYBRID_KERNEL) Code Format Check #241: Pull request #732 synchronize by Xuan-1998
Action required Xuan-1998:unordered-kernels
Xuan-1998:unordered-kernels
Action required
fix: get_nvlink_gbs may return 0 as a divisor Code Format Check #240: Pull request #740 synchronize by cxxlink
Action required cxxlink:fix_div_0
cxxlink:fix_div_0
Action required
fix: st.bulk support for size operand with 32-bit length is introduce… Code Format Check #239: Pull request #737 synchronize by cxxlink
Action required cxxlink:main
cxxlink:main
Action required
fix: get_nvlink_gbs may return 0 as a divisor Code Format Check #238: Pull request #740 opened by cxxlink
Action required cxxlink:fix_div_0
cxxlink:fix_div_0
Action required
fix: st.bulk support for size operand with 32-bit length is introduce… Code Format Check #236: Pull request #737 synchronize by cxxlink
Action required cxxlink:main
cxxlink:main
Action required
Add unordered hybrid kernels for transports without ordered delivery (opt-in via EP_HYBRID_KERNEL) Code Format Check #235: Pull request #732 synchronize by Xuan-1998
Action required Xuan-1998:unordered-kernels
Xuan-1998:unordered-kernels
Action required
Add unordered hybrid kernels for transports without ordered delivery (opt-in via EP_HYBRID_KERNEL) Code Format Check #234: Pull request #732 synchronize by Xuan-1998
Action required Xuan-1998:unordered-kernels
Xuan-1998:unordered-kernels
Action required
fix: support system NCCL installations Code Format Check #233: Pull request #738 opened by janbernloehr
Action required janbernloehr:system-nccl-main
janbernloehr:system-nccl-main
Action required
fix: st.bulk support for size operand with 32-bit length is introduce… Code Format Check #232: Pull request #737 synchronize by cxxlink
Action required cxxlink:main
cxxlink:main
Action required
fix: st.bulk support for size operand with 32-bit length is introduce… Code Format Check #231: Pull request #737 opened by cxxlink
Action required cxxlink:main
cxxlink:main
Action required
bug fix: probe RDMA link rate via sysfs, survive probe failure Code Format Check #230: Pull request #734 synchronize by Xuan-1998
Action required Xuan-1998:envs-sysfs-rdma-probe
Xuan-1998:envs-sysfs-rdma-probe
Action required
Add unordered hybrid kernels for transports without ordered delivery (opt-in via EP_HYBRID_KERNEL) Code Format Check #229: Pull request #732 synchronize by Xuan-1998
Action required Xuan-1998:unordered-kernels
Xuan-1998:unordered-kernels
Action required
fix: return default_value from get_env<int> on unparsable env values Code Format Check #228: Pull request #736 opened by shoemoney
Action required shoemoney:fix/get-env-uninitialized
shoemoney:fix/get-env-uninitialized
Action required
Add unordered hybrid kernels for transports without ordered delivery (opt-in via EP_HYBRID_KERNEL) Code Format Check #227: Pull request #732 synchronize by Xuan-1998
Action required Xuan-1998:unordered-kernels
Xuan-1998:unordered-kernels
Action required
tests: bounded pressure loops and teardown barrier for test_ep Code Format Check #226: Pull request #735 opened by Xuan-1998
Action required Xuan-1998:test-ep-pressure-iterations
Xuan-1998:test-ep-pressure-iterations
Action required
bug fix: probe RDMA link rate via sysfs, survive probe failure Code Format Check #225: Pull request #734 opened by Xuan-1998
Action required Xuan-1998:envs-sysfs-rdma-probe
Xuan-1998:envs-sysfs-rdma-probe
Action required
Add unordered hybrid kernels for transports without ordered delivery (opt-in via EP_HYBRID_KERNEL) Code Format Check #224: Pull request #732 opened by Xuan-1998
Action required Xuan-1998:unordered-kernels
Xuan-1998:unordered-kernels
Action required
utils: only reuse PyTorch's NCCL comm when _comm_ptr() is non-null Code Format Check #223: Pull request #727 synchronize by dmvevents
Action required dmvevents:fix/null-guard-reused-comm-ptr
dmvevents:fix/null-guard-reused-comm-ptr
Action required
feat(low_latency): support top-16 routing and hidden-dim 3584 Code Format Check #222: Pull request #731 opened by samsja
Action required samsja:fix/k3-topk16-hidden3584
samsja:fix/k3-topk16-hidden3584
Action required
[Runtime] Preflight full intranode P2P matrix Code Format Check #221: Pull request #730 synchronize by 0z5a
Action required 0z5a:fix/p2p-preflight
0z5a:fix/p2p-preflight
Action required
[Runtime] Preflight full intranode P2P matrix Code Format Check #220: Pull request #730 opened by 0z5a
Action required 0z5a:fix/p2p-preflight
0z5a:fix/p2p-preflight
Action required
docs: fix cached dispatch outputs Code Format Check #219: Pull request #729 opened by bvolpato
Action required bvolpato:bvolpato/fix-cached-dispatch-readme
bvolpato:bvolpato/fix-cached-dispatch-readme
Action required