🔴 [VLM] Add base model without head (#37033)

* i guessreverted all CdGen classes

* style

* llava onevision

* fix copies

* fix some tests

* some more tests

* dump

* skip these

* nevermind, i am dumb

* revert fix not needed

* fixup

* fixup

* another fixup

* more fixup to make ci finally happy

* fixup after rebasing

* fix qwen tests

* add internVL + typos here and there

* image token index -> id

* style

* fix init weights

* revert blip-2 not supported

* address comments

* fix copies

* revert blip2 test file as well

* as discussed internally, revert back CdGen models

* fix some tests

* fix more tests for compile

* CI red

* fix copies

* enumerate explicitly allowed models

* address comments

* fix tests

* fixup

* style again

* add tests for new model class

* another fixup ( x _ x )

* [fixup] unused attributes can be removed post-deprecation

tests/models/aria/test_modeling_aria.py

@@ -21,6 +21,7 @@ import requests
 from transformers import (
     AriaConfig,
     AriaForConditionalGeneration,
+    AriaModel,
     AriaTextConfig,
     AutoProcessor,
     AutoTokenizer,
@@ -175,7 +176,7 @@ class AriaForConditionalGenerationModelTest(ModelTesterMixin, GenerationTesterMi
     Model tester for `AriaForConditionalGeneration`.
     """
 
-    all_model_classes = (AriaForConditionalGeneration,) if is_torch_available() else ()
+    all_model_classes = (AriaModel, AriaForConditionalGeneration) if is_torch_available() else ()
     test_pruning = False
     test_head_masking = False
     _is_composite = True
@@ -281,6 +282,18 @@ class AriaForConditionalGenerationModelTest(ModelTesterMixin, GenerationTesterMi
     def test_generate_from_inputs_embeds_with_static_cache(self):
         pass
 
+    @unittest.skip(reason="Aria uses nn.MHA which is not compatible with offloading")
+    def test_cpu_offload(self):
+        pass
+
+    @unittest.skip(reason="Aria uses nn.MHA which is not compatible with offloading")
+    def test_disk_offload_bin(self):
+        pass
+
+    @unittest.skip(reason="Aria uses nn.MHA which is not compatible with offloading")
+    def test_disk_offload_safetensors(self):
+        pass
+
 
 @require_torch
 class AriaForConditionalGenerationIntegrationTest(unittest.TestCase):