[#13534][chore] AutoDeploy: Remove Two Model Speculative Decoding Support (#13532)

Signed-off-by: Govind Ramnarayan <105831528+govind-ramnarayan@users.noreply.github.com>

Author: govind-ramnarayan

tensorrt_llm/_torch/auto_deploy/llm_args.py

@@ -109,35 +109,57 @@ def ensure_no_custom_parallel_config(cls, value: Any, info: ValidationInfo) -> A
         msg = "AutoDeploy only supports parallelization via the `world_size` argument."
         return _check_for_default_value_only(cls, value, info, msg)
 
+    @model_validator(mode="after")
+    def validate_supported_speculative_config(self):
+        spec_config = self.speculative_config
+        if spec_config is None:
+            return self
+
+        if isinstance(spec_config, MTPDecodingConfig):
+            if not spec_config.mtp_eagle_one_model or spec_config.use_mtp_vanilla:
+                raise ValueError(
+                    "AutoDeploy only supports MTP speculative decoding with "
+                    "mtp_eagle_one_model=True and use_mtp_vanilla=False "
+                    f"(got mtp_eagle_one_model={spec_config.mtp_eagle_one_model}, "
+                    f"use_mtp_vanilla={spec_config.use_mtp_vanilla})."
+                )
+        elif isinstance(spec_config, EagleDecodingConfig):
+            if not spec_config.eagle3_one_model:
+                raise ValueError(
+                    "AutoDeploy only supports Eagle speculative decoding with "
+                    f"eagle3_one_model=True (got eagle3_one_model={spec_config.eagle3_one_model})."
+                )
+        else:
+            raise ValueError(
+                "AutoDeploy only supports speculative decoding via "
+                "MTPDecodingConfig(mtp_eagle_one_model=True) or "
+                "EagleDecodingConfig(eagle3_one_model=True)."
+            )
+
+        self.model_factory = "eagle_one_model"
+        return self
+
     @model_validator(mode="after")
     def setup_hidden_state_capture(self):
         spec_config = self.speculative_config
         if spec_config is None:
             return self
 
         if isinstance(spec_config, MTPDecodingConfig):
-            if not spec_config.mtp_eagle_one_model:
-                return self
-            if spec_config.use_mtp_vanilla:
-                raise ValueError("mtp_eagle_one_model and use_mtp_vanilla cannot both be enabled")
             if spec_config.max_draft_len is None:
                 raise ValueError(
                     "MTPDecodingConfig.max_draft_len must not be None when mtp_eagle_one_model is "
                     "enabled. Ensure num_nextn_predict_layers is set in the model config."
                 )
             capture_layers = {-1}
-            self.model_factory = "eagle_one_model"
-        elif isinstance(spec_config, EagleDecodingConfig):
+        else:
+            assert isinstance(spec_config, EagleDecodingConfig)
             if spec_config.max_draft_len is None:
                 raise ValueError(
                     "EagleDecodingConfig.max_draft_len must not be None. "
                     "Provide a positive integer for max_draft_len."
                 )
             capture_layers = spec_config.eagle3_layers_to_capture
-            if spec_config.eagle3_one_model:
-                self.model_factory = "eagle_one_model"
-        else:
-            return self
 
         self.transforms["detect_hidden_states_for_capture"]["enabled"] = True
         self.transforms["detect_hidden_states_for_capture"]["eagle3_layers_to_capture"] = (
@@ -223,11 +245,6 @@ def validate_and_init_tokenizer(self):
 
     device: str = Field(default="cuda", description="The device to use for the model.", frozen=True)
 
-    draft_checkpoint_loader: Optional[object] = Field(
-        default=None,
-        description="The checkpoint loader to use for the draft model when using speculative decoding with two models.",
-    )
-
     ### INFERENCE OPTIMIZER CONFIG #################################################################
     mode: Literal["graph", "transformers"] = Field(
         default="graph",