@@ -531,7 +531,8 @@ def gather_from_tp_ranks(self, tensor: torch.Tensor) -> List[torch.Tensor]:
         torch.distributed.all_gather(gathered, tensor, group=self.tp_group)
         return gathered
 
-    def _count_wildcard_groups(self, pattern: str) -> int:
+    @staticmethod
+    def _count_wildcard_groups(pattern: str) -> int:
         """Count the number of wildcard capture groups in a pattern.
 
         Args:
 
@@ -166,9 +166,6 @@ class MambaModelProvider(TransformerConfig, ModelProviderMixin[MCoreMambaModel])
     mamba_stack_spec: Union[ModuleSpec, Callable[[], ModuleSpec], Callable[["MambaModelProvider"], ModuleSpec]] = (
         get_default_mamba_stack_spec
     )
-    mtp_mamba_stack_spec: Union[ModuleSpec, Callable[[], ModuleSpec], Callable[["MambaModelProvider"], ModuleSpec]] = (
-        get_default_mamba_stack_spec
-    )
     vocab_size: Optional[int] = None
     should_pad_vocab: bool = False
     hf_model_id: Optional[str] = None
 
@@ -37,17 +37,6 @@
 logger = logging.getLogger(__name__)
 
 
-def _count_wildcards(pattern: str) -> int:
-    # Mirrors MegatronParamMapping._count_wildcard_groups (private there).
-    count = 0
-    remaining = pattern
-    while "**" in remaining:
-        count += 1
-        remaining = remaining.replace("**", "", 1)
-    count += remaining.count("*")
-    return count
-
-
 def _replace_wildcards(pattern: str, captures: Tuple[str, ...]) -> str:
     """Replace ** then * sequentially with captures."""
     out = pattern
@@ -108,8 +97,10 @@ def __init__(
 
         # Precompute wildcard counts to disambiguate resolve() calls coming from
         # megatron_to_hf_lookup vs hf_to_megatron_lookup.
-        self._megatron_wc = _count_wildcards(self.megatron_param)
-        self._hf_wc = _count_wildcards(self.hf_param) if isinstance(self.hf_param, str) else 0
+        self._megatron_wc = MegatronParamMapping._count_wildcard_groups(self.megatron_param)
+        self._hf_wc = (
+            MegatronParamMapping._count_wildcard_groups(self.hf_param) if isinstance(self.hf_param, str) else 0
+        )
 
     def resolve(self, captures: Tuple[str, ...]) -> MegatronParamMapping:
         # We primarily expect captures from Megatron pattern matching.
@@ -161,7 +152,9 @@ def resolve(self, captures: Tuple[str, ...]) -> MegatronParamMapping:
 
         return AutoMapping(megatron_param=resolved_megatron, hf_param=resolved_hf)
 
-    # These are never called: this mapping always resolves into an AutoMapping.
+    # Required by ABC but never called at runtime: resolve() always returns an
+    # AutoMapping, so hf_to_megatron/megatron_to_hf are invoked on that resolved
+    # instance, not on this class.
     def hf_to_megatron(self, hf_weights: torch.Tensor, megatron_module: torch.nn.Module) -> torch.Tensor:
         raise NotImplementedError
 
@@ -189,8 +182,8 @@ def __init__(
         self._mtp_layers_per_block = int(mtp_layers_per_block)
         if self._mtp_layers_per_block <= 0:
             raise ValueError("mtp_layers_per_block must be > 0 for MTP flattening QKV mapping.")
-        self._megatron_wc = _count_wildcards(self.megatron_param)
-        self._hf_wc = _count_wildcards(q)  # q/k/v share pattern structure here
+        self._megatron_wc = MegatronParamMapping._count_wildcard_groups(self.megatron_param)
+        self._hf_wc = MegatronParamMapping._count_wildcard_groups(q)  # q/k/v share pattern structure here
 
     def resolve(self, captures: Tuple[str, ...]) -> MegatronParamMapping:
         # Expect captures from Megatron lookup: (outer, inner)
@@ -207,6 +200,9 @@ def resolve(self, captures: Tuple[str, ...]) -> MegatronParamMapping:
 
         return QKVMapping(megatron_param=resolved_megatron, q=resolved_q, k=resolved_k, v=resolved_v)
 
+    # Required by ABC but never called at runtime: resolve() always returns a
+    # QKVMapping, so hf_to_megatron/megatron_to_hf are invoked on that resolved
+    # instance, not on this class.
     def hf_to_megatron(self, hf_weights: Dict[str, torch.Tensor], megatron_module: torch.nn.Module) -> torch.Tensor:
         raise NotImplementedError
 
 
@@ -167,36 +167,6 @@ class Nemotron3NanoProvider(NemotronHModelProvider):
     moe_router_group_topk: int = 1
 
 
-@dataclass
-class Nemotron3SuperProvider(NemotronHModelProvider):
-    """Configuration for a 3B parameter Nemotron 3 Super model."""
-
-    seq_length: int = 262144
-    num_query_groups: int = 2
-    hybrid_layer_pattern: str = (
-        "MEMEMEM*EMEMEMEM*EMEMEMEM*EMEMEMEMEM*EMEMEMEMEM*EMEMEMEMEM*EMEMEMEMEM*EMEMEMEM*EMEMEMEME"
-    )
-    hidden_size: int = 4096
-    mamba_num_heads: int = 128
-    kv_channels: int = 128
-    mamba_state_dim: int = 128
-    ffn_hidden_size: int = 2688
-    num_attention_heads: int = 32
-    mamba_head_dim: int = 64
-    num_moe_experts: int = 512
-    moe_ffn_hidden_size: int = 2688
-    moe_shared_expert_intermediate_size: int = 5376
-    moe_router_topk: int = 22
-    moe_router_topk_scaling_factor: float = 5.0
-    moe_router_num_groups: int = 1
-    moe_router_group_topk: int = 1
-    moe_latent_size: int = 1024
-    moe_shared_expert_overlap: bool = False
-    mtp_num_layers: int = 2
-    mtp_hybrid_override_pattern: str = "*E"
-    mtp_use_repeated_layer: bool = False
-
-
 # -----------------------------------------------------------------------------
 # Deprecated aliases (to be removed in a future release)
 # -----------------------------------------------------------------------------
 
@@ -17,7 +17,6 @@
 
 from megatron.bridge.models.nemotronh.nemotron_h_provider import (
     Nemotron3NanoProvider,
-    Nemotron3SuperProvider,
     NemotronHModel4BProvider,
     NemotronHModel8BProvider,
     NemotronHModel47BProvider,
@@ -284,10 +283,6 @@ def test_nemotron_nano_12b_v2_inherits_from_base(self):
         """Test Nemotron Nano v2 12B provider inherits from NemotronHModelProvider."""
         assert issubclass(NemotronNano12Bv2Provider, NemotronHModelProvider)
 
-    def test_nemotron_3_super_inherits_from_base(self):
-        """Test Nemotron 3 Super provider inherits from NemotronHModelProvider."""
-        assert issubclass(Nemotron3SuperProvider, NemotronHModelProvider)
-
     def test_provide_method_inherited(self):
         """Test that provide method works correctly in inherited classes."""
         # Test with Nemotron-H 4B
@@ -299,7 +294,6 @@ def test_provide_method_inherited(self):
             NemotronNano9Bv2Provider(),
             NemotronNano12Bv2Provider(),
             Nemotron3NanoProvider(),
-            Nemotron3SuperProvider(),
         ]
 
         for provider in providers:
@@ -463,93 +457,6 @@ def test_nemotron_3_nano_inherits_moe_defaults(self):
         assert provider.moe_shared_expert_overlap is True
 
 
-class TestNemotron3SuperProvider:
-    """Test cases for Nemotron3SuperProvider class."""
-
-    def test_nemotron_3_super_default_configuration(self):
-        """Test Nemotron 3 Super model has correct default configuration."""
-        provider = Nemotron3SuperProvider()
-
-        # Check Nemotron 3 Super specific configuration
-        assert provider.seq_length == 262144
-        assert provider.hidden_size == 4096
-        assert provider.num_attention_heads == 32
-        assert provider.num_query_groups == 2
-        assert provider.mamba_num_heads == 128
-        assert provider.kv_channels == 128
-        assert provider.mamba_state_dim == 128
-        assert provider.ffn_hidden_size == 2688
-        assert provider.mamba_head_dim == 64
-        assert (
-            provider.hybrid_layer_pattern
-            == "MEMEMEM*EMEMEMEM*EMEMEMEM*EMEMEMEMEM*EMEMEMEMEM*EMEMEMEMEM*EMEMEMEMEM*EMEMEMEM*EMEMEMEME"
-        )
-        # num_layers is derived from hybrid_layer_pattern in finalize()
-        assert provider.num_layers is None
-        assert len(provider.hybrid_layer_pattern) == 88
-
-    def test_nemotron_3_super_moe_configuration(self):
-        """Test Nemotron 3 Super model MoE-specific configuration."""
-        provider = Nemotron3SuperProvider()
-
-        # Check MoE-specific configuration
-        assert provider.num_moe_experts == 512
-        assert provider.moe_ffn_hidden_size == 2688
-        assert provider.moe_shared_expert_intermediate_size == 5376
-        assert provider.moe_router_topk == 22
-        assert provider.moe_router_topk_scaling_factor == 5.0
-        assert provider.moe_router_num_groups == 1
-        assert provider.moe_router_group_topk == 1
-        assert provider.moe_latent_size == 1024
-        assert provider.moe_shared_expert_overlap is False
-
-    def test_nemotron_3_super_mtp_configuration(self):
-        """Test Nemotron 3 Super model MTP-specific configuration."""
-        provider = Nemotron3SuperProvider()
-
-        # Check MTP-specific configuration
-        assert provider.mtp_num_layers == 2
-        assert provider.mtp_hybrid_override_pattern == "*E"
-        assert provider.mtp_use_repeated_layer is False
-
-    def test_nemotron_3_super_override_configuration(self):
-        """Test Nemotron 3 Super model with overridden configuration."""
-        provider = Nemotron3SuperProvider(
-            seq_length=16384,
-            hidden_dropout=0.1,
-            num_moe_experts=64,
-        )
-
-        # Check overridden values
-        assert provider.seq_length == 16384
-        assert provider.hidden_dropout == 0.1
-        assert provider.num_moe_experts == 64
-
-        # Check critical defaults remain (num_layers derived from pattern in finalize())
-        assert provider.num_layers is None
-        assert len(provider.hybrid_layer_pattern) == 88
-        assert provider.hidden_size == 4096
-        assert provider.mamba_num_heads == 128
-
-    def test_nemotron_3_super_inherits_from_base(self):
-        """Test Nemotron 3 Super provider inherits from NemotronHModelProvider."""
-        assert issubclass(Nemotron3SuperProvider, NemotronHModelProvider)
-
-    def test_nemotron_3_super_inherits_moe_defaults(self):
-        """Test Nemotron 3 Super inherits MoE defaults from base class."""
-        provider = Nemotron3SuperProvider()
-
-        # Check inherited MoE defaults from NemotronHModelProvider
-        assert provider.moe_aux_loss_coeff == 0.0001
-        assert provider.moe_router_score_function == "sigmoid"
-        assert provider.moe_router_enable_expert_bias is True
-        assert provider.moe_router_load_balancing_type == "seq_aux_loss"
-        assert provider.moe_router_dtype == "fp32"
-        assert provider.moe_grouped_gemm is True
-        assert provider.moe_token_dispatcher_type == "alltoall"
-        assert provider.moe_permute_fusion is True
-
-
 class TestHybridPatterns:
     """Test hybrid override patterns of Nemotron-H providers."""
 
@@ -563,7 +470,6 @@ def test_hybrid_patterns_contain_mamba_and_attention(self):
             NemotronNano9Bv2Provider(),
             NemotronNano12Bv2Provider(),
             Nemotron3NanoProvider(),
-            Nemotron3SuperProvider(),
         ]
 
         for provider in providers:
 
@@ -311,7 +311,7 @@ def test_moe_model_configuration(self, recipe_fn):
         """Test MoE-specific model configuration from provider."""
         config = recipe_fn()
 
-        # Check MoE settings from Nemotron3SuperProvider
+        # Check MoE settings from AutoBridge provider
         assert config.model.num_moe_experts == 512
         assert config.model.moe_ffn_hidden_size == 2688
         assert config.model.moe_shared_expert_intermediate_size == 5376