Binyang Li 205556f8cb Merge feature/ep benchmark routing updates ...

Resolve the C++ benchmark conflict by combining NCCL-EP-style random top-k routing and masked selections with the current BF16/FP8 MoERuntime API and format-aware byte accounting.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Copilot-Session: efbacae6-f679-430b-bc16-b45ae162fc76

2026-07-13 21:49:11 +00:00

..

CMakeLists.txt

Update EP benchmarks for current low-latency API

2026-07-13 21:45:58 +00:00

cupti_kernel_timer.cpp

Update EP benchmarks for current low-latency API

2026-07-13 21:45:58 +00:00

ep_bench_ll.py

Update EP benchmarks for current low-latency API

2026-07-13 21:45:58 +00:00

mscclpp_ep_bench.cu

Merge feature/ep benchmark routing updates

2026-07-13 21:49:11 +00:00

run_ep_bench.py

Update EP benchmarks for current low-latency API

2026-07-13 21:45:58 +00:00

test_internode_multirank.py

Binyli/ep revise (#828)

2026-07-06 21:14:29 -07:00

test_intranode_multirank.py

Binyli/ep revise (#828)

2026-07-06 21:14:29 -07:00

test_low_latency_multirank.py

Add configurable FP8 low-latency dispatch

2026-07-13 18:47:21 +00:00