EngineX-Ascend/enginex-ascend-910-llama.cpp
10
0
Fork 0
Code Issues Pull Requests Actions 4 Projects Releases Wiki Activity

model : add BailingMoeV2 support (#16063)

Browse Source 
* add BailingMoeV2 support

* update llm types

* undo

* undo

* update llm types

* add model collection link

* update

* almost working

* correct group selection and rename n_group_exp

* avoid large top_k and use argmax instead for now

if we had something like argmax2 that would be equivalent, but this works fine until then

* poke

* skip group selection when there are no tokens

* fix 1T conversion

* hopefully fixed expert group selection

third time's the charm?

* make expert group selection generally available

The new LLaDA2Moe model uses this method too, make it generally available regardless of architecture.

* allow n_expert_groups to be 1 (Kimi K2)

* address review suggestions
This commit is contained in:
Sigbjørn Skjæret
2025-10-20 21:38:20 +02:00
committed by GitHub
parent c9c1972e2c
commit 84bf3c6778
15 changed files with 521 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
gguf-py/gguf/gguf_writer.py
View File

@@ -755,6 +755,12 @@ class GGUFWriter:
def add_expert_shared_count(self, count: int) -> None:
self.add_uint32(Keys.LLM.EXPERT_SHARED_COUNT.format(arch=self.arch), count)
def add_expert_group_count(self, count: int) -> None:
self.add_uint32(Keys.LLM.EXPERT_GROUP_COUNT.format(arch=self.arch), count)
def add_expert_group_used_count(self, count: int) -> None:
self.add_uint32(Keys.LLM.EXPERT_GROUP_USED_COUNT.format(arch=self.arch), count)
def add_expert_weights_scale(self, value: float) -> None:
self.add_float32(Keys.LLM.EXPERT_WEIGHTS_SCALE.format(arch=self.arch), value)