vllm/vllm/distributed
2024-10-04 16:43:50 -07:00
..
device_communicators [torch.compile] improve allreduce registration (#9061) 2024-10-04 16:43:50 -07:00
__init__.py [Core][Refactor] move parallel_utils into vllm/distributed (#3950) 2024-04-10 15:33:30 -07:00
communication_op.py [Bugfix] Fix weight loading for Chameleon when TP>1 (#7410) 2024-08-13 05:33:41 +00:00
parallel_state.py [torch.compile] improve allreduce registration (#9061) 2024-10-04 16:43:50 -07:00
utils.py [MISC] Introduce pipeline parallelism partition strategies (#6920) 2024-07-31 12:02:17 -07:00