EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

Update model_loader deps and qqq quantization deps (#2220) (#2318)

Browse Source 
Co-authored-by: HandH1998 <1335248067@qq.com>
This commit is contained in:
Yineng Zhang
2024-12-02 23:22:13 +08:00
committed by GitHub
parent 33deca81b5
commit 85e1a6f3aa
58 changed files with 2363 additions and 366 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
python/sglang/srt/models/torch_native_llama.py
View File

@@ -52,7 +52,6 @@ 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.activation import SiluAndMul
from sglang.srt.layers.layernorm import RMSNorm
@@ -66,6 +65,7 @@ from sglang.srt.layers.vocab_parallel_embedding import (
)
from sglang.srt.managers.schedule_batch import global_server_args_dict
from sglang.srt.model_executor.forward_batch_info import ForwardBatch
from sglang.srt.model_loader.weight_utils import default_weight_loader
tp_size = get_tensor_model_parallel_world_size()
tp_rank = get_tensor_model_parallel_rank()
@@ -388,7 +388,6 @@ class TorchNativeLlamaForCausalLM(nn.Module):
self,
config: LlamaConfig,
quant_config: Optional[QuantizationConfig] = None,
cache_config=None,
) -> None:
super().__init__()
self.config = config