Upload 2025-11-27/ci_results_run_models_gpu/new_failures.json with huggingface_hub

2025-11-27/ci_results_run_models_gpu/new_failures.json

@@ -0,0 +1,28 @@
+{
+    "glm4": {
+        "single-gpu": [
+            "tests/models/glm4/test_modeling_glm4.py::Glm4ModelTest::test_flash_attn_2_equivalence"
+        ],
+        "multi-gpu": []
+    },
+    "hunyuan_v1_moe": {
+        "single-gpu": [],
+        "multi-gpu": [
+            "tests/models/hunyuan_v1_moe/test_modeling_hunyuan_v1_moe.py::HunYuanMoEV1ModelTest::test_flash_attn_2_equivalence"
+        ]
+    },
+    "idefics2": {
+        "single-gpu": [
+            "tests/models/idefics2/test_modeling_idefics2.py::Idefics2ForConditionalGenerationIntegrationTest::test_flash_attn_2_eager_equivalence"
+        ],
+        "multi-gpu": [
+            "tests/models/idefics2/test_modeling_idefics2.py::Idefics2ForConditionalGenerationIntegrationTest::test_flash_attn_2_eager_equivalence"
+        ]
+    },
+    "roberta_prelayernorm": {
+        "single-gpu": [],
+        "multi-gpu": [
+            "tests/models/roberta_prelayernorm/test_modeling_roberta_prelayernorm.py::RobertaPreLayerNormModelTest::test_flash_attn_2_inference_equivalence_right_padding"
+        ]
+    }
+}