ver4a/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

llama : support BailingMoE (Ling) (#12634)

Browse source
This commit is contained in:
Sigbjørn Skjæret 2025-03-30 22:21:03 +02:00 • committed by GitHub
parent 4663bd353c
commit 2c3f8b850a
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
13 changed files with 404 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

24
gguf-py/gguf/constants.py
View file

@ -287,6 +287,7 @@ class MODEL_ARCH(IntEnum):
CHAMELEON = auto()
WAVTOKENIZER_DEC = auto()
PLM = auto()
BAILINGMOE = auto()
class MODEL_TENSOR(IntEnum):
@ -490,6 +491,7 @@ MODEL_ARCH_NAMES: dict[MODEL_ARCH, str] = {
MODEL_ARCH.CHAMELEON: "chameleon",
MODEL_ARCH.WAVTOKENIZER_DEC: "wavtokenizer-dec",
MODEL_ARCH.PLM: "plm",
MODEL_ARCH.BAILINGMOE: "bailingmoe",
}
TENSOR_NAMES: dict[MODEL_TENSOR, str] = {
@ -1667,6 +1669,25 @@ MODEL_TENSORS: dict[MODEL_ARCH, list[MODEL_TENSOR]] = {
MODEL_TENSOR.POSNET_ATTN_V,
MODEL_TENSOR.POSNET_ATTN_OUT,
],
MODEL_ARCH.BAILINGMOE: [
MODEL_TENSOR.TOKEN_EMBD,
MODEL_TENSOR.OUTPUT_NORM,
MODEL_TENSOR.OUTPUT,
MODEL_TENSOR.ROPE_FREQS,
MODEL_TENSOR.ATTN_NORM,
MODEL_TENSOR.ATTN_Q,
MODEL_TENSOR.ATTN_K,
MODEL_TENSOR.ATTN_V,
MODEL_TENSOR.ATTN_OUT,
MODEL_TENSOR.FFN_GATE_INP,
MODEL_TENSOR.FFN_NORM,
MODEL_TENSOR.FFN_GATE_EXP,
MODEL_TENSOR.FFN_DOWN_EXP,
MODEL_TENSOR.FFN_UP_EXP,
MODEL_TENSOR.FFN_GATE_SHEXP,
MODEL_TENSOR.FFN_DOWN_SHEXP,
MODEL_TENSOR.FFN_UP_SHEXP,
],
# TODO
}
@ -1719,6 +1740,9 @@ MODEL_TENSOR_SKIP: dict[MODEL_ARCH, list[MODEL_TENSOR]] = {
MODEL_TENSOR.ROPE_FREQS,
MODEL_TENSOR.ATTN_ROT_EMBD,
],
MODEL_ARCH.BAILINGMOE: [
MODEL_TENSOR.ROPE_FREQS,
],
}
#

1
gguf-py/gguf/tensor_mapping.py
View file

@ -29,6 +29,7 @@ class TensorNameMap:
"shared", # t5
"rwkv.embeddings", # rwkv6
"model.embeddings", # rwkv7
"model.word_embeddings", # bailingmoe
),
# Token type embeddings