Skip to content

Commit 06a0165

Browse files
authored
Merge pull request #725 from ROCm/355_wip_moe
Support fp8 with static scales
2 parents e8b970d + c867c0f commit 06a0165

File tree

8 files changed

+1726
-411
lines changed

8 files changed

+1726
-411
lines changed

vllm/config/parallel.py

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -29,6 +29,7 @@
2929

3030
logger = init_logger(__name__)
3131

32+
ExpertPlacementStrategy = Literal["linear", "round_robin"]
3233
DistributedExecutorBackend = Literal["ray", "mp", "uni", "external_launcher"]
3334

3435

0 commit comments

Comments
 (0)