Rename triton_fused_moe -> fused_moe_triton (#2163)

python/sglang/srt/models/dbrx.py

@@ -28,6 +28,7 @@ from vllm.model_executor.layers.rotary_embedding import get_rope
 from vllm.model_executor.model_loader.weight_utils import default_weight_loader
 from vllm.transformers_utils.configs.dbrx import DbrxConfig
 
+from sglang.srt.layers.fused_moe_triton import fused_moe
 from sglang.srt.layers.linear import (
     QKVParallelLinear,
     ReplicatedLinear,
@@ -36,7 +37,6 @@ from sglang.srt.layers.linear import (
 from sglang.srt.layers.logits_processor import LogitsProcessor
 from sglang.srt.layers.quantization.base_config import QuantizationConfig
 from sglang.srt.layers.radix_attention import RadixAttention
-from sglang.srt.layers.triton_fused_moe import fused_moe
 from sglang.srt.layers.vocab_parallel_embedding import (
     DEFAULT_VOCAB_PADDING_SIZE,
     ParallelLMHead,

python/sglang/srt/models/deepseek.py

@@ -30,6 +30,7 @@ from vllm.model_executor.layers.rotary_embedding import get_rope
 from vllm.model_executor.model_loader.weight_utils import default_weight_loader
 
 from sglang.srt.layers.activation import SiluAndMul
+from sglang.srt.layers.fused_moe_triton import fused_moe
 from sglang.srt.layers.layernorm import RMSNorm
 from sglang.srt.layers.linear import (
     MergedColumnParallelLinear,
@@ -40,7 +41,6 @@ from sglang.srt.layers.linear import (
 from sglang.srt.layers.logits_processor import LogitsProcessor
 from sglang.srt.layers.quantization.base_config import QuantizationConfig
 from sglang.srt.layers.radix_attention import RadixAttention
-from sglang.srt.layers.triton_fused_moe import fused_moe
 from sglang.srt.layers.vocab_parallel_embedding import (
     ParallelLMHead,
     VocabParallelEmbedding,

python/sglang/srt/models/deepseek_v2.py

@@ -31,6 +31,7 @@ from vllm.model_executor.layers.rotary_embedding import get_rope
 from vllm.model_executor.model_loader.weight_utils import default_weight_loader
 
 from sglang.srt.layers.activation import SiluAndMul
+from sglang.srt.layers.fused_moe_triton import FusedMoE
 from sglang.srt.layers.layernorm import RMSNorm
 from sglang.srt.layers.linear import (
     ColumnParallelLinear,
@@ -41,7 +42,6 @@ from sglang.srt.layers.linear import (
 from sglang.srt.layers.logits_processor import LogitsProcessor
 from sglang.srt.layers.quantization.base_config import QuantizationConfig
 from sglang.srt.layers.radix_attention import RadixAttention
-from sglang.srt.layers.triton_fused_moe import FusedMoE
 from sglang.srt.layers.vocab_parallel_embedding import (
     ParallelLMHead,
     VocabParallelEmbedding,

python/sglang/srt/models/grok.py

@@ -31,7 +31,7 @@ from vllm.model_executor.layers.rotary_embedding import get_rope
 from vllm.model_executor.model_loader.loader import DefaultModelLoader
 from vllm.model_executor.model_loader.weight_utils import default_weight_loader
 
-from sglang.srt.layers.fused_moe import FusedMoE
+from sglang.srt.layers.fused_moe_grok import FusedMoE
 from sglang.srt.layers.layernorm import RMSNorm
 from sglang.srt.layers.linear import (
     QKVParallelLinear,

python/sglang/srt/models/mixtral.py

@@ -25,6 +25,7 @@ from vllm.distributed import get_tensor_model_parallel_world_size
 from vllm.model_executor.layers.rotary_embedding import get_rope
 from vllm.model_executor.model_loader.weight_utils import default_weight_loader
 
+from sglang.srt.layers.fused_moe_triton import FusedMoE
 from sglang.srt.layers.layernorm import RMSNorm
 from sglang.srt.layers.linear import (
     QKVParallelLinear,
@@ -35,7 +36,6 @@ from sglang.srt.layers.logits_processor import LogitsProcessor
 from sglang.srt.layers.quantization.base_config import QuantizationConfig
 from sglang.srt.layers.radix_attention import RadixAttention
 from sglang.srt.layers.torchao_utils import apply_torchao_config_
-from sglang.srt.layers.triton_fused_moe import FusedMoE
 from sglang.srt.layers.vocab_parallel_embedding import (
     ParallelLMHead,
     VocabParallelEmbedding,

python/sglang/srt/models/olmoe.py

@@ -38,11 +38,11 @@ from vllm.model_executor.model_loader.weight_utils import default_weight_loader
 from vllm.utils import print_warning_once
 
 from sglang.srt.layers.activation import SiluAndMul
+from sglang.srt.layers.fused_moe_triton import FusedMoE
 from sglang.srt.layers.layernorm import RMSNorm
 from sglang.srt.layers.logits_processor import LogitsProcessor, LogitsProcessorOutput
 from sglang.srt.layers.quantization.base_config import QuantizationConfig
 from sglang.srt.layers.radix_attention import RadixAttention
-from sglang.srt.layers.triton_fused_moe import FusedMoE
 from sglang.srt.layers.vocab_parallel_embedding import (
     ParallelLMHead,
     VocabParallelEmbedding,

python/sglang/srt/models/qwen2_moe.py

@@ -30,6 +30,7 @@ from vllm.model_executor.layers.rotary_embedding import get_rope
 from vllm.model_executor.model_loader.weight_utils import default_weight_loader
 
 from sglang.srt.layers.activation import SiluAndMul
+from sglang.srt.layers.fused_moe_triton import FusedMoE
 from sglang.srt.layers.layernorm import RMSNorm
 from sglang.srt.layers.linear import (
     MergedColumnParallelLinear,
@@ -41,7 +42,6 @@ from sglang.srt.layers.logits_processor import LogitsProcessor
 from sglang.srt.layers.quantization.base_config import QuantizationConfig
 from sglang.srt.layers.radix_attention import RadixAttention
 from sglang.srt.layers.torchao_utils import apply_torchao_config_
-from sglang.srt.layers.triton_fused_moe import FusedMoE
 from sglang.srt.layers.vocab_parallel_embedding import (
     ParallelLMHead,
     VocabParallelEmbedding,

python/sglang/srt/models/xverse_moe.py

@@ -34,10 +34,10 @@ from vllm.model_executor.layers.linear import (
 from vllm.model_executor.layers.rotary_embedding import get_rope
 from vllm.model_executor.model_loader.weight_utils import default_weight_loader
 
+from sglang.srt.layers.fused_moe_triton import fused_moe
 from sglang.srt.layers.logits_processor import LogitsProcessor
 from sglang.srt.layers.quantization.base_config import QuantizationConfig
 from sglang.srt.layers.radix_attention import RadixAttention
-from sglang.srt.layers.triton_fused_moe import fused_moe
 from sglang.srt.layers.vocab_parallel_embedding import (
     ParallelLMHead,
     VocabParallelEmbedding,