karylab_agents/vllm
1
0
Fork 0
forked from Karylab-cklius/vllm
Code Pull Requests 2 Actions 1 Packages Activity

Allow FlashInfer A2A backends for TRTLLM FP8 MoE Modular (#46661)

Browse Source 
Signed-off-by: Thien Tran <gau.nernst@yahoo.com.sg>
This commit is contained in:
Thien Tran
2026-07-08 14:58:39 -07:00
committed by GitHub
parent 49abadaedb
commit 0d2f4e7c9c
1 changed files with 2 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
vllm/model_executor/layers/fused_moe/experts/trtllm_fp8_moe.py
+2
View File
@@ -136,6 +136,8 @@ class TrtLlmFp8ExpertsModular(TrtLlmFp8ExpertsBase, mk.FusedMoEExpertsModular):
not moe_parallel_config.use_all2all_kernels
or moe_parallel_config.use_ag_rs_all2all_kernels
or moe_parallel_config.use_deepep_v2_kernels
or moe_parallel_config.use_fi_nvl_one_sided_kernels
or moe_parallel_config.use_fi_nvl_two_sided_kernels
) and not moe_parallel_config.enable_eplb
@staticmethod