fix: allow lookup of models registered at runtime

adds tests for ModelRegistryHelper to stabilize behavior

llama_stack/providers/datatypes.py

@@ -22,6 +22,27 @@ from llama_stack.schema_utils import json_schema_type
 
 
 class ModelsProtocolPrivate(Protocol):
+    """
+    Protocol for model management.
+
+    This allows users to register their preferred model identifiers.
+
+    Model registration requires -
+     - a provider, used to route the registration request
+     - a model identifier, user's intended name for the model during inference
+     - a provider model identifier, a model identifier supported by the provider
+
+    Providers will only accept registration for provider model ids they support.
+
+    Example,
+      register: provider x my-model-id x provider-model-id
+       -> Error if provider does not support provider-model-id
+       -> Error if my-model-id is already registered
+       -> Success if provider supports provider-model-id
+      inference: my-model-id x ...
+       -> Provider uses provider-model-id for inference
+    """
+
     async def register_model(self, model: Model) -> Model: ...
 
     async def unregister_model(self, model_id: str) -> None: ...

llama_stack/providers/utils/inference/model_registry.py

@@ -59,6 +59,8 @@ def build_model_entry(provider_model_id: str, model_descriptor: str) -> Provider
 
 class ModelRegistryHelper(ModelsProtocolPrivate):
     def __init__(self, model_entries: List[ProviderModelEntry]):
+        self.supported_model_ids = {entry.provider_model_id for entry in model_entries}
+
         self.alias_to_provider_id_map = {}
         self.provider_id_to_llama_model_map = {}
         for entry in model_entries:
@@ -79,6 +81,16 @@ class ModelRegistryHelper(ModelsProtocolPrivate):
         return self.provider_id_to_llama_model_map.get(provider_model_id, None)
 
     async def register_model(self, model: Model) -> Model:
+        if model.provider_resource_id not in self.supported_model_ids:
+            raise ValueError(
+                f"Model id '{model.provider_resource_id}' is not supported. Supported ids are: {', '.join(self.supported_model_ids)}"
+            )
+        if model.model_id in self.alias_to_provider_id_map:
+            # be idemopotent
+            if model.provider_resource_id != self.alias_to_provider_id_map[model.model_id]:
+                raise ValueError(
+                    f"Model id '{model.model_id}' is already registered. Please use a different id or unregister it first."
+                )
         if model.model_type == ModelType.embedding:
             # embedding models are always registered by their provider model id and does not need to be mapped to a llama model
             provider_resource_id = model.provider_resource_id
@@ -108,7 +120,12 @@ class ModelRegistryHelper(ModelsProtocolPrivate):
                     ALL_HUGGINGFACE_REPOS_TO_MODEL_DESCRIPTOR[llama_model]
                 )
 
+        self.alias_to_provider_id_map[model.model_id] = model.provider_resource_id
+
         return model
 
     async def unregister_model(self, model_id: str) -> None:
-        pass
+        if model_id not in self.alias_to_provider_id_map:
+            raise ValueError(f"Model id '{model_id}' is not registered.")
+
+        del self.alias_to_provider_id_map[model_id]