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

[NVIDIA] Support flashinfer Mega Moe (#31470)

Browse Source 
Co-authored-by: djns99 <40156487+djns99@users.noreply.github.com>
Co-authored-by: 云挚 <ningyunxiao.nyx@antgroup.com>
Co-authored-by: Yangmin Li <yangminl@nvidia.com>
Co-authored-by: Po-Han Huang (NVIDIA) <53919306+nvpohanh@users.noreply.github.com>
This commit is contained in:
Shu Wang
2026-09-10 00:22:47 -07:00
committed by GitHub
co-authored by djns99 云挚 Yangmin Li Po-Han Huang
parent c0b790cf7f
commit 1b77f498a0
31 changed files with 2021 additions and 72 deletions
Download Patch File Download Diff File Expand all files Collapse all files
python/sglang/srt/runtime_context.py
+1
View File
@@ -622,6 +622,7 @@ class Resources(_FlagGroupBase):
# Persistent reusable CUDA events for non-EP DP TBO, keyed by
# (kind, subbatch) — see dp_attention._tbo_event for why reuse matters.
tbo_event_pool: dict = msgspec.field(default_factory=dict)
flashinfer_megamoe_workspaces: dict = msgspec.field(default_factory=dict)
# State capturers (installed by their subsystems when capture is on).
indexer_capturer: Any = None
experts_capturer: Any = None