support llava video (#426)

python/sglang/srt/models/mixtral.py

@@ -8,34 +8,28 @@ import torch
 import torch.nn.functional as F
 from torch import nn
 from transformers import MixtralConfig
+from vllm.distributed import (
+    get_tensor_model_parallel_rank,
+    get_tensor_model_parallel_world_size,
+    tensor_model_parallel_all_reduce,
+)
 from vllm.model_executor.layers.layernorm import RMSNorm
 from vllm.model_executor.layers.linear import (
     QKVParallelLinear,
     ReplicatedLinear,
     RowParallelLinear,
 )
-from vllm.model_executor.layers.quantization.base_config import (
-    QuantizationConfig)
+from vllm.model_executor.layers.quantization.base_config import QuantizationConfig
 from vllm.model_executor.layers.rotary_embedding import get_rope
 from vllm.model_executor.layers.vocab_parallel_embedding import (
     ParallelLMHead,
     VocabParallelEmbedding,
 )
-from vllm.distributed import (
-    tensor_model_parallel_all_reduce,
-)
-from vllm.distributed import (
-    get_tensor_model_parallel_rank,
-    get_tensor_model_parallel_world_size,
-)
-from sglang.srt.weight_utils import (
-    default_weight_loader,
-    hf_model_weights_iterator,
-)
 
 from sglang.srt.layers.logits_processor import LogitsProcessor
 from sglang.srt.layers.radix_attention import RadixAttention
 from sglang.srt.managers.router.model_runner import InputMetadata
+from sglang.srt.weight_utils import default_weight_loader, hf_model_weights_iterator
 
 
 class MixtralMLP(nn.Module):