several fixes

llama_stack/providers/remote/inference/bedrock/models.py

@@ -4,7 +4,7 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack.models.llama.datatypes import CoreModelId
+from llama_stack.models.llama.sku_types import CoreModelId
 from llama_stack.providers.utils.inference.model_registry import (
     build_hf_repo_model_entry,
 )

llama_stack/providers/remote/inference/cerebras/cerebras.py

@@ -28,8 +28,8 @@ from llama_stack.apis.inference import (
     ToolConfig,
     ToolDefinition,
     ToolPromptFormat,
+    TopKSamplingStrategy,
 )
-from llama_stack.models.llama.datatypes import TopKSamplingStrategy
 from llama_stack.providers.utils.inference.model_registry import (
     ModelRegistryHelper,
 )

llama_stack/providers/remote/inference/cerebras/models.py

@@ -4,7 +4,7 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack.models.llama.datatypes import CoreModelId
+from llama_stack.models.llama.sku_types import CoreModelId
 from llama_stack.providers.utils.inference.model_registry import (
     build_hf_repo_model_entry,
 )

llama_stack/providers/remote/inference/databricks/databricks.py

@@ -28,7 +28,7 @@ from llama_stack.apis.inference import (
     ToolDefinition,
     ToolPromptFormat,
 )
-from llama_stack.models.llama.datatypes import CoreModelId
+from llama_stack.models.llama.sku_types import CoreModelId
 from llama_stack.providers.utils.inference.model_registry import (
     ModelRegistryHelper,
     build_hf_repo_model_entry,

llama_stack/providers/remote/inference/fireworks/models.py

@@ -5,7 +5,7 @@
 # the root directory of this source tree.
 
 from llama_stack.apis.models.models import ModelType
-from llama_stack.models.llama.datatypes import CoreModelId
+from llama_stack.models.llama.sku_types import CoreModelId
 from llama_stack.providers.utils.inference.model_registry import (
     ProviderModelEntry,
     build_hf_repo_model_entry,

llama_stack/providers/remote/inference/nvidia/models.py

@@ -5,7 +5,7 @@
 # the root directory of this source tree.
 
 from llama_stack.apis.models import ModelType
-from llama_stack.models.llama.datatypes import CoreModelId
+from llama_stack.models.llama.sku_types import CoreModelId
 from llama_stack.providers.utils.inference.model_registry import (
     ProviderModelEntry,
     build_hf_repo_model_entry,

llama_stack/providers/remote/inference/nvidia/nvidia.py

@@ -29,15 +29,13 @@ from llama_stack.apis.inference import (
     LogProbConfig,
     Message,
     ResponseFormat,
+    SamplingParams,
     TextTruncation,
     ToolChoice,
     ToolConfig,
-)
-from llama_stack.models.llama.datatypes import (
-    SamplingParams,
     ToolDefinition,
-    ToolPromptFormat,
 )
+from llama_stack.models.llama.datatypes import ToolPromptFormat
 from llama_stack.providers.utils.inference.model_registry import (
     ModelRegistryHelper,
 )

llama_stack/providers/remote/inference/nvidia/openai_utils.py

@@ -19,11 +19,9 @@ from llama_stack.apis.inference import (
     CompletionRequest,
     CompletionResponse,
     CompletionResponseStreamChunk,
+    GreedySamplingStrategy,
     JsonSchemaResponseFormat,
     TokenLogProbs,
-)
-from llama_stack.models.llama.datatypes import (
-    GreedySamplingStrategy,
     TopKSamplingStrategy,
     TopPSamplingStrategy,
 )

llama_stack/providers/remote/inference/ollama/models.py

@@ -5,7 +5,7 @@
 # the root directory of this source tree.
 
 from llama_stack.apis.models.models import ModelType
-from llama_stack.models.llama.datatypes import CoreModelId
+from llama_stack.models.llama.sku_types import CoreModelId
 from llama_stack.providers.utils.inference.model_registry import (
     ProviderModelEntry,
     build_hf_repo_model_entry,

llama_stack/providers/remote/inference/sambanova/models.py

@@ -4,7 +4,7 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack.models.llama.datatypes import CoreModelId
+from llama_stack.models.llama.sku_types import CoreModelId
 from llama_stack.providers.utils.inference.model_registry import (
     build_hf_repo_model_entry,
 )

llama_stack/providers/remote/inference/sambanova/sambanova.py

@@ -21,6 +21,7 @@ from llama_stack.apis.inference import (
     CompletionMessage,
     EmbeddingsResponse,
     EmbeddingTaskType,
+    GreedySamplingStrategy,
     Inference,
     LogProbConfig,
     Message,
@@ -35,12 +36,9 @@ from llama_stack.apis.inference import (
     ToolDefinition,
     ToolPromptFormat,
     ToolResponseMessage,
-    UserMessage,
-)
-from llama_stack.models.llama.datatypes import (
-    GreedySamplingStrategy,
     TopKSamplingStrategy,
     TopPSamplingStrategy,
+    UserMessage,
 )
 from llama_stack.providers.utils.inference.model_registry import ModelRegistryHelper
 from llama_stack.providers.utils.inference.openai_compat import (

llama_stack/providers/remote/inference/together/models.py

@@ -5,7 +5,7 @@
 # the root directory of this source tree.
 
 from llama_stack.apis.models.models import ModelType
-from llama_stack.models.llama.datatypes import CoreModelId
+from llama_stack.models.llama.sku_types import CoreModelId
 from llama_stack.providers.utils.inference.model_registry import (
     ProviderModelEntry,
     build_hf_repo_model_entry,

llama_stack/providers/remote/inference/vllm/vllm.py

@@ -96,7 +96,6 @@ def _convert_to_vllm_tool_calls_in_response(
             call_id=call.id,
             tool_name=call.function.name,
             arguments=json.loads(call.function.arguments),
-            arguments_json=call.function.arguments,
         )
         for call in tool_calls
     ]
@@ -176,7 +175,6 @@ async def _process_vllm_chat_completion_stream_response(
                                 call_id=tool_call_buf.call_id,
                                 tool_name=tool_call_buf.tool_name,
                                 arguments=args,
-                                arguments_json=args_str,
                             ),
                             parse_status=ToolCallParseStatus.succeeded,
                         ),

llama_stack/providers/remote/post_training/nvidia/models.py

@@ -6,7 +6,7 @@
 
 from typing import List
 
-from llama_stack.models.llama.datatypes import CoreModelId
+from llama_stack.models.llama.sku_types import CoreModelId
 from llama_stack.providers.utils.inference.model_registry import (
     ProviderModelEntry,
     build_hf_repo_model_entry,