Upload 2025-12-01/ci_results_run_models_gpu/new_failures.json with huggingface_hub

2025-12-01/ci_results_run_models_gpu/new_failures.json

@@ -0,0 +1,34 @@
+{
+    "chameleon": {
+        "single-gpu": [
+            "tests/models/chameleon/test_modeling_chameleon.py::ChameleonVision2SeqModelTest::test_flash_attention_2_padding_matches_padding_free_with_position_ids_and_fa_kwargs"
+        ],
+        "multi-gpu": []
+    },
+    "gemma3n": {
+        "single-gpu": [
+            "tests/models/gemma3n/test_modeling_gemma3n.py::Gemma3nTextModelTest::test_flash_attention_2_padding_matches_padding_free_with_position_ids_and_fa_kwargs"
+        ],
+        "multi-gpu": []
+    },
+    "gpt_neo": {
+        "single-gpu": [
+            "tests/models/gpt_neo/test_modeling_gpt_neo.py::GPTNeoModelTest::test_eager_matches_fa2_generate"
+        ],
+        "multi-gpu": []
+    },
+    "hunyuan_v1_moe": {
+        "single-gpu": [],
+        "multi-gpu": [
+            "tests/models/hunyuan_v1_moe/test_modeling_hunyuan_v1_moe.py::HunYuanMoEV1ModelTest::test_flash_attention_2_padding_matches_padding_free_with_position_ids_and_fa_kwargs"
+        ]
+    },
+    "roberta": {
+        "single-gpu": [
+            "tests/models/roberta/test_modeling_roberta.py::RobertaModelTest::test_flash_attn_2_inference_equivalence"
+        ],
+        "multi-gpu": [
+            "tests/models/roberta/test_modeling_roberta.py::RobertaModelTest::test_flash_attn_2_inference_equivalence_right_padding"
+        ]
+    }
+}