chenchenghao/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

Support DeepSeek V3.2 Exp (#11061)

Browse Source 
Co-authored-by: Stefan He <11166516+hebiao064@users.noreply.github.com>
Co-authored-by: Liangsheng Yin <95566987+hnyls2002@users.noreply.github.com>
Co-authored-by: Baizhou Zhang <56809903+fridge003@users.noreply.github.com>
Co-authored-by: DarkSharpness <76582120+darksharpness@users.noreply.github.com>
Co-authored-by: ZhengdQin <46387172+zhengdqin@users.noreply.github.com>
Co-authored-by: DarkSharpness <2040703891@qq.com>
Co-authored-by: hnyls2002 <lsyincs@gmail.com>
Co-authored-by: Zhengda Qin <zhengdqin@gmail.com>
Co-authored-by: Liangsheng Yin <hnyls2002@gmail.com>
Co-authored-by: HAI <hixiao@gmail.com>
Co-authored-by: Baizhou Zhang <sobereddiezhang@gmail.com>
This commit is contained in:
fzyzcjy
2025-10-06 15:24:15 +08:00
committed by GitHub
parent 292a867ad9
commit efbc687c28
29 changed files with 4540 additions and 139 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
python/sglang/srt/two_batch_overlap.py
View File

@@ -705,6 +705,8 @@ class TboForwardBatchPreparer:
extend_num_tokens=extend_num_tokens,
attn_backend=output_attn_backend,
num_token_non_padded=out_num_token_non_padded,
# TODO: handle it when we need TBO + DeepSeek V3.2
num_token_non_padded_cpu=None,
tbo_split_seq_index=None,
tbo_parent_token_range=(start_token_index, end_token_index),
tbo_children=None,