biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Refactor] Replace activation: str with MoEActivation enum (#33843)

Browse Source 
Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
This commit is contained in:
Michael Goin
2026-02-11 20:29:32 -05:00
committed by GitHub
parent 83b47f67b1
commit ff1f83b056
48 changed files with 474 additions and 282 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
tests/kernels/moe/test_cutlass_moe.py
View File

@@ -12,6 +12,7 @@ from tests.kernels.moe.utils import make_dummy_moe_config
from vllm import _custom_ops as ops
from vllm.config import ParallelConfig, VllmConfig, set_current_vllm_config
from vllm.model_executor.layers.fused_moe import fused_experts, fused_topk
from vllm.model_executor.layers.fused_moe.activation import MoEActivation
from vllm.model_executor.layers.fused_moe.config import (
FUSED_MOE_UNQUANTIZED_CONFIG,
FusedMoEQuantConfig,
@@ -531,7 +532,7 @@ def test_run_cutlass_moe_fp8(
c_strides1 = torch.full((e,), 2 * n, device="cuda", dtype=torch.int64)
c_strides2 = torch.full((e,), k, device="cuda", dtype=torch.int64)
activation = "silu"
activation = MoEActivation.SILU
a1q, a1q_scale = moe_kernel_quantize_input(
mt.a, mt.a_scale, torch.float8_e4m3fn, per_act_token
)