xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-16 10:45:45 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Bench] Add qwen-next in benchmark_moe.py (#24661)

Browse Source 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
This commit is contained in:
Jee Jee Li 2025-09-11 21:29:43 +08:00 committed by GitHub
parent f510715882
commit d11ec124a0
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 5 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
benchmarks/kernels/benchmark_moe.py
View File

@ -594,7 +594,11 @@ def main(args: argparse.Namespace):
E = config.n_routed_experts E = config.n_routed_experts
topk = config.num_experts_per_tok topk = config.num_experts_per_tok
intermediate_size = config.moe_intermediate_size intermediate_size = config.moe_intermediate_size
elif config.architectures[0] in ("Qwen2MoeForCausalLM", "Qwen3MoeForCausalLM"): elif config.architectures[0] in (
"Qwen2MoeForCausalLM",
"Qwen3MoeForCausalLM",
"Qwen3NextForCausalLM",
):
E = config.num_experts E = config.num_experts
topk = config.num_experts_per_tok topk = config.num_experts_per_tok
intermediate_size = config.moe_intermediate_size intermediate_size = config.moe_intermediate_size