Support compressed tensors fp8w8a8 (#4743)

This commit is contained in:
Xiaoyu Zhang
2025-03-27 04:21:25 +08:00
committed by GitHub
parent 45fdf1f7f3
commit 04e3ff6975
30 changed files with 2386 additions and 113 deletions

View File

@@ -1316,7 +1316,10 @@ vllm_get_world_group = None
def monkey_patch_vllm_parallel_state(reverse: bool = False):
import vllm.distributed.parallel_state as vllm_parrlel_state
try:
import vllm.distributed.parallel_state as vllm_parrlel_state
except ImportError:
return
global vllm_get_pp_group, vllm_get_tp_group, vllm_get_world_group
if vllm_get_pp_group is None: