youkaichao 3de6e6a30e

[core][distributed] support n layers % pp size != 0 (#6115)

2024-07-03 16:40:31 -07:00

..

device_communicators

[misc][cuda] use nvml to avoid accidentally cuda initialization (#6007)

2024-06-30 20:07:34 -07:00

__init__.py

[Core][Refactor] move parallel_utils into vllm/distributed (#3950)

2024-04-10 15:33:30 -07:00

communication_op.py

[Core][Distributed] code deduplication in tp&pp with coordinator(#5293)

2024-06-12 17:27:08 -07:00

parallel_state.py

[core][distributed] custom allreduce when pp size > 1 (#6117)

2024-07-03 14:41:32 -07:00

utils.py

[core][distributed] support n layers % pp size != 0 (#6115)

2024-07-03 16:40:31 -07:00