cleanup models unused import 2/n (#2952)

python/sglang/srt/models/gpt2.py

@@ -17,7 +17,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 """Inference-only GPT-2 model compatible with HuggingFace weights."""
-from typing import Iterable, List, Optional, Tuple
+from typing import Iterable, Optional, Tuple
 
 import torch
 from torch import nn

python/sglang/srt/models/olmo.py

@@ -15,7 +15,7 @@
 # Adapted from
 # https://github.com/vllm-project/vllm/blob/c7f2cf2b7f67bce5842fedfdba508440fe257375/vllm/model_executor/models/olmo.py#L1
 """Inference-only OLMo model compatible with HuggingFace weights."""
-from typing import Iterable, List, Optional, Tuple
+from typing import Iterable, Optional, Tuple
 
 import torch
 from torch import nn

python/sglang/srt/models/olmoe.py

@@ -17,26 +17,21 @@
 
 """Inference-only OLMoE model compatible with HuggingFace weights."""
 
-from typing import Any, Dict, Iterable, List, Optional, Tuple
+from typing import Any, Dict, Iterable, Optional, Tuple
 
 import torch
-import torch.nn.functional as F
 from torch import nn
 from transformers import PretrainedConfig
 from vllm.model_executor.layers.rotary_embedding import get_rope
 
-from sglang.srt.distributed import (
+from sglang.srt.distributed import get_tensor_model_parallel_world_size
-    get_tensor_model_parallel_world_size,
-    tensor_model_parallel_all_reduce,
-)
-from sglang.srt.layers.activation import SiluAndMul
 from sglang.srt.layers.layernorm import RMSNorm
 from sglang.srt.layers.linear import (
     QKVParallelLinear,
     ReplicatedLinear,
     RowParallelLinear,
 )
-from sglang.srt.layers.logits_processor import LogitsProcessor, LogitsProcessorOutput
+from sglang.srt.layers.logits_processor import LogitsProcessor
 from sglang.srt.layers.moe.fused_moe_triton import FusedMoE
 from sglang.srt.layers.quantization.base_config import QuantizationConfig
 from sglang.srt.layers.radix_attention import RadixAttention