chenchenghao/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 11 Packages Projects Releases Wiki Activity

Reorg moe code (#2563)

Browse Source
This commit is contained in:
Ke Bao
2024-12-24 01:10:22 +08:00
committed by GitHub
parent 23e5e50fd5
commit e835a50021
88 changed files with 338 additions and 344 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
python/sglang/srt/model_executor/cuda_graph_runner.py
View File

@@ -25,12 +25,12 @@ from vllm.distributed import get_tensor_model_parallel_rank
from vllm.distributed.parallel_state import graph_capture
from vllm.model_executor.custom_op import CustomOp
from sglang.srt.layers.fused_moe_patch import fused_moe_forward_native
from sglang.srt.layers.logits_processor import (
LogitsMetadata,
LogitsProcessor,
LogitsProcessorOutput,
)
from sglang.srt.layers.moe.fused_moe_native import fused_moe_forward_native
from sglang.srt.model_executor.forward_batch_info import ForwardBatch, ForwardMode
from sglang.srt.utils import maybe_torch_compile, monkey_patch_vllm_all_gather