Unverified Commit a1d5bc4c authored by Elfie Guo's avatar Elfie Guo Committed by GitHub
Browse files

Avoid using flashinfer_allreduce_fusion when dp attention is enabled. (#11632)

parent a8023891
......@@ -337,6 +337,7 @@ class LayerCommunicator:
static_conditions_met = (
(not self.is_last_layer)
and (self._context.tp_size > 1)
and not is_dp_attention_enabled()
and get_global_server_args().enable_flashinfer_allreduce_fusion
and _is_flashinfer_available
)
......
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