Temporarily disable OOMing llama vision decoder test (#6908)

Temp disable OOMing llama vision decoder test

Author: jackzhxng

.ci/scripts/gather_test_models.py

@@ -25,7 +25,7 @@
         "resnet50": "linux.12xlarge",
         "llava": "linux.12xlarge",
         "llama3_2_vision_encoder": "linux.12xlarge",
-        "llama3_2_text_decoder": "linux.12xlarge",
+        # "llama3_2_text_decoder": "linux.12xlarge",  # TODO: re-enable test when Huy's change is in / model gets smaller.
         # This one causes timeout on smaller runner, the root cause is unclear (T161064121)
         "dl3": "linux.12xlarge",
         "emformer_join": "linux.12xlarge",

examples/models/__init__.py

@@ -19,7 +19,7 @@
     "llama2": ("llama", "Llama2Model"),
     "llama": ("llama", "Llama2Model"),
     "llama3_2_vision_encoder": ("llama3_2_vision", "FlamingoVisionEncoderModel"),
-    "llama3_2_text_decoder": ("llama3_2_vision", "Llama3_2Decoder"),
+    # "llama3_2_text_decoder": ("llama3_2_vision", "Llama3_2Decoder"),
     "lstm": ("lstm", "LSTMModel"),
     "mobilebert": ("mobilebert", "MobileBertModelExample"),
     "mv2": ("mobilenet_v2", "MV2Model"),