Skip to content

GitLab

Unverified Commit 7e4e709b authored by Alexander Matveev's avatar Alexander Matveev Committed by GitHub
Browse files

[V1] TPU - Fix fused MOE (#15834) 


Signed-off-by: default avatarAlexander Matveev <amatveev@redhat.com>
parent 63d8eabe
Hide whitespace changes
Inline Side-by-side
Showing with 1 addition and 1 deletion
vllm/model_executor/layers/fused_moe/layer.py
View file @ 7e4e709b
...@@ -309,7 +309,7 @@ class UnquantizedFusedMoEMethod(FusedMoEMethodBase, CustomOp): ...@@ -309,7 +309,7 @@ class UnquantizedFusedMoEMethod(FusedMoEMethodBase, CustomOp):
expert_map=expert_map, expert_map=expert_map,
renormalize=renormalize) renormalize=renormalize)
forward_native = forward_cuda forward_native = forward_tpu if current_platform.is_tpu else forward_cuda
def determine_expert_map( def determine_expert_map(
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment