[Perf] Wire silu_and_mul_per_block_quant into TritonFP8MoE (MiniMax-M2) (#42497)

Signed-off-by: qianlihuang <yiliu.dong@qq.com>
Signed-off-by: Yiliu Dong <91178480+qianlihuang@users.noreply.github.com>
Co-authored-by: qianlihuang <yiliu.dong@qq.com>
This commit is contained in:
Yiliu Dong
2026-05-17 21:57:04 -04:00
committed by GitHub
co-authored by qianlihuang
parent 966903eb93
commit 03ddc1c9bc
@@ -31,6 +31,9 @@ from vllm.model_executor.layers.fused_moe.utils import (
_resize_cache,
moe_kernel_quantize_input,
)
from vllm.model_executor.layers.quantization.utils.fp8_utils import (
is_deep_gemm_e8m0_used,
)
from vllm.model_executor.layers.quantization.utils.quant_utils import (
QuantKey,
kFp8Dynamic128Sym,
@@ -283,20 +286,36 @@ class TritonExperts(LoRAExpertsMixin, mk.FusedMoEExpertsModular):
top_k_num=top_k_num,
)
self.activation(
activation, intermediate_cache2, intermediate_cache1.view(-1, N)
)
a2q_scale: torch.Tensor | None = None
qintermediate_cache2, a2q_scale = moe_kernel_quantize_input(
intermediate_cache2,
a2_scale,
self.quant_dtype,
self.per_act_token_quant,
self.block_shape,
quantization_emulation=self.quantization_emulation,
)
# Fuse SiLU+Mul + FP8 block quantize into a single kernel
# when conditions permit (gated SiLU, fp8 block quant with
# group_size=128, no LoRA requiring the BF16 intermediate).
if (
activation == MoEActivation.SILU
and self.quant_config.use_fp8_w8a8
and self.block_shape == [128, 128]
and lora_context is None
and not is_deep_gemm_e8m0_used()
):
qintermediate_cache2, a2q_scale = ops.silu_and_mul_per_block_quant(
intermediate_cache1.view(-1, N),
group_size=128,
quant_dtype=current_platform.fp8_dtype(),
)
else:
self.activation(
activation, intermediate_cache2, intermediate_cache1.view(-1, N)
)
qintermediate_cache2, a2q_scale = moe_kernel_quantize_input(
intermediate_cache2,
a2_scale,
self.quant_dtype,
self.per_act_token_quant,
self.block_shape,
quantization_emulation=self.quantization_emulation,
)
invoke_fused_moe_triton_kernel(
qintermediate_cache2,