tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-29 20:45:04 +00:00
Code Issues Packages Projects Releases Wiki Activity

model : add support for ERNIE 4.5 0.3B model (#14408)

Browse Source 
Add Day-0 support for Baidu ERNIE 4.5 0.3B model.

Signed-off-by: Weizhao Ouyang <weizhao.ouyang@arm.com>
This commit is contained in:
Weizhao Ouyang
2025-06-28 22:08:21 +08:00
committed by GitHub
parent b25e92774e
commit 566c16fcce
6 changed files with 260 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
gguf-py/gguf/constants.py
View File

@ -354,6 +354,7 @@ class MODEL_ARCH(IntEnum):
BAILINGMOE = auto()
DOTS1 = auto()
ARCEE = auto()
ERNIE4_5 = auto()
class VISION_PROJECTOR_TYPE(IntEnum):
@ -654,6 +655,7 @@ MODEL_ARCH_NAMES: dict[MODEL_ARCH, str] = {
MODEL_ARCH.BAILINGMOE: "bailingmoe",
MODEL_ARCH.DOTS1: "dots1",
MODEL_ARCH.ARCEE: "arcee",
MODEL_ARCH.ERNIE4_5: "ernie4_5",
}
VISION_PROJECTOR_TYPE_NAMES: dict[VISION_PROJECTOR_TYPE, str] = {
@ -2177,6 +2179,20 @@ MODEL_TENSORS: dict[MODEL_ARCH, list[MODEL_TENSOR]] = {
MODEL_TENSOR.FFN_DOWN,
MODEL_TENSOR.FFN_UP,
],
MODEL_ARCH.ERNIE4_5: [
MODEL_TENSOR.TOKEN_EMBD,
MODEL_TENSOR.OUTPUT_NORM,
MODEL_TENSOR.OUTPUT,
MODEL_TENSOR.ATTN_NORM,
MODEL_TENSOR.ATTN_Q,
MODEL_TENSOR.ATTN_K,
MODEL_TENSOR.ATTN_V,
MODEL_TENSOR.ATTN_OUT,
MODEL_TENSOR.FFN_NORM,
MODEL_TENSOR.FFN_GATE,
MODEL_TENSOR.FFN_DOWN,
MODEL_TENSOR.FFN_UP,
],
# TODO
}