feat(transformers): add VaultGemma (v4.57.1)

mindone/transformers/__init__.py

@@ -1365,6 +1365,7 @@
 )
 from .models.univnet import UnivNetModel
 from .models.upernet import UperNetForSemanticSegmentation, UperNetPreTrainedModel
+from .models.vaultgemma import VaultGemmaForCausalLM, VaultGemmaModel, VaultGemmaPreTrainedModel
 from .models.video_llava import VideoLlavaForConditionalGeneration, VideoLlavaPreTrainedModel
 from .models.videomae import (
     VideoMAEForPreTraining,

mindone/transformers/models/__init__.py

@@ -233,6 +233,7 @@
     unispeech_sat,
     univnet,
     upernet,
+    vaultgemma,
     video_llava,
     videomae,
     vilt,

mindone/transformers/models/auto/configuration_auto.py

@@ -262,6 +262,7 @@
         ("unispeech-sat", "UniSpeechSatConfig"),
         ("univnet", "UnivNetConfig"),
         ("upernet", "UperNetConfig"),
+        ("vaultgemma", "VaultGemmaConfig"),
         ("video_llava", "VideoLlavaConfig"),
         ("videomae", "VideoMAEConfig"),
         ("vilt", "ViltConfig"),
@@ -534,6 +535,7 @@
         ("unispeech-sat", "UniSpeechSat"),
         ("univnet", "UnivNet"),
         ("upernet", "UPerNet"),
+        ("vaultgemma", "VaultGemma"),
         ("video_llava", "VideoLlava"),
         ("videomae", "VideoMAE"),
         ("vilt", "ViLT"),

mindone/transformers/models/auto/modeling_auto.py

@@ -237,6 +237,7 @@
         ("unispeech", "UniSpeechModel"),
         ("unispeech-sat", "UniSpeechSatModel"),
         ("univnet", "UnivNetModel"),
+        ("vaultgemma", "VaultGemmaModel"),
         ("videomae", "VideoMAEModel"),
         ("vilt", "ViltModel"),
         ("visual_bert", "VisualBertModel"),
@@ -489,6 +490,7 @@
         ("stablelm", "StableLmForCausalLM"),
         ("starcoder2", "Starcoder2ForCausalLM"),
         ("trocr", "TrOCRForCausalLM"),
+        ("vaultgemma", "VaultGemmaForCausalLM"),
         ("whisper", "WhisperForCausalLM"),
         ("xglm", "XGLMForCausalLM"),
         ("xlm", "XLMWithLMHeadModel"),

mindone/transformers/models/vaultgemma/__init__.py

@@ -0,0 +1,18 @@
+# coding=utf-8
+# Copyright 2025 the HuggingFace Team. All rights reserved.
+#
+# This code is adapted from https://github.com/huggingface/transformers
+# with modifications to run transformers on mindspore.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from .modeling_vaultgemma import *