Skip to content

GitLab

Unverified Commit 67f3fb08 authored by Jee Jee Li's avatar Jee Jee Li Committed by GitHub
Browse files

[Bench] Add DeepSeekV32 to MoE benchmark (#25962) 


Signed-off-by: default avatarJee Jee Li <pandaleefree@gmail.com>
parent 43b752c3
Hide whitespace changes
Inline Side-by-side
Showing with 2 additions and 1 deletion
benchmarks/kernels/benchmark_moe.py
View file @ 67f3fb08
...@@ -584,8 +584,9 @@ def main(args: argparse.Namespace): ...@@ -584,8 +584,9 @@ def main(args: argparse.Namespace):
topk = config.num_experts_per_tok topk = config.num_experts_per_tok
intermediate_size = config.intermediate_size intermediate_size = config.intermediate_size
elif config.architectures[0] in ( elif config.architectures[0] in (
"DeepseekV3ForCausalLM",
"DeepseekV2ForCausalLM", "DeepseekV2ForCausalLM",
"DeepseekV3ForCausalLM",
"DeepseekV32ForCausalLM",
"Glm4MoeForCausalLM", "Glm4MoeForCausalLM",
): ):
E = config.n_routed_experts E = config.n_routed_experts
......
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