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

Support data parallelism (static) (#480)

Browse Source 
Co-authored-by: Ying Sheng <ying.sheng@databricks.com>
Co-authored-by: Lianmin Zheng <lianminzheng@gmail.com>
Co-authored-by: Liangsheng Yin <hnyls2002@gmail.com>
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
This commit is contained in:
Ying Sheng
2024-05-27 21:24:10 -07:00
committed by GitHub
parent 565d727409
commit 0463f7fb52
32 changed files with 580 additions and 181 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
python/sglang/srt/models/llava.py
View File

@@ -10,8 +10,8 @@ from transformers.models.llava.modeling_llava import LlavaMultiModalProjector
from vllm.model_executor.layers.quantization.base_config import QuantizationConfig
from vllm.model_executor.model_loader.weight_utils import default_weight_loader
from sglang.srt.managers.router.infer_batch import ForwardMode
from sglang.srt.managers.router.model_runner import InputMetadata
from sglang.srt.managers.controller.infer_batch import ForwardMode
from sglang.srt.managers.controller.model_runner import InputMetadata
from sglang.srt.mm_utils import (
get_anyres_image_grid_shape,
unpad_image,