423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2026-02-28 16:56:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

[model] support MiniCPM-o-4.5 (#10163)

Browse Source 
Co-authored-by: Yaowei Zheng <hiyouga@buaa.edu.cn>
This commit is contained in:
Hertz
2026-02-04 23:21:27 +08:00
committed by GitHub
parent 1a02717fa8
commit 8bedfafa4e
7 changed files with 22 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
src/llamafactory/extras/constants.py
View File

@@ -14,7 +14,7 @@
import os
from collections import OrderedDict, defaultdict
from enum import Enum, unique
from enum import StrEnum, unique
from peft.utils import SAFETENSORS_WEIGHTS_NAME as SAFE_ADAPTER_WEIGHTS_NAME
from peft.utils import WEIGHTS_NAME as ADAPTER_WEIGHTS_NAME
@@ -110,7 +110,7 @@ V_HEAD_WEIGHTS_NAME = "value_head.bin"
V_HEAD_SAFE_WEIGHTS_NAME = "value_head.safetensors"
class AttentionFunction(str, Enum):
class AttentionFunction(StrEnum):
AUTO = "auto"
DISABLED = "disabled"
SDPA = "sdpa"
@@ -118,21 +118,21 @@ class AttentionFunction(str, Enum):
FA3 = "fa3"
class EngineName(str, Enum):
class EngineName(StrEnum):
HF = "huggingface"
VLLM = "vllm"
SGLANG = "sglang"
KT = "ktransformers"
class DownloadSource(str, Enum):
class DownloadSource(StrEnum):
DEFAULT = "hf"
MODELSCOPE = "ms"
OPENMIND = "om"
@unique
class QuantizationMethod(str, Enum):
class QuantizationMethod(StrEnum):
r"""Borrowed from `transformers.utils.quantization_config.QuantizationMethod`."""
BNB = "bnb"
@@ -146,7 +146,7 @@ class QuantizationMethod(str, Enum):
FP8 = "fp8"
class RopeScaling(str, Enum):
class RopeScaling(StrEnum):
LINEAR = "linear"
DYNAMIC = "dynamic"
YARN = "yarn"
@@ -1840,6 +1840,10 @@ register_model_group(
DownloadSource.DEFAULT: "openbmb/MiniCPM-o-2_6",
DownloadSource.MODELSCOPE: "OpenBMB/MiniCPM-o-2_6",
},
"MiniCPM-o-4_5": {
DownloadSource.DEFAULT: "openbmb/MiniCPM-o-4_5",
DownloadSource.MODELSCOPE: "OpenBMB/MiniCPM-o-4_5",
},
},
template="minicpm_o",
multimodal=True,