pre-commit fixes

llama_stack/providers/utils/inference/litellm_openai_mixin.py

@@ -8,7 +8,6 @@ from typing import AsyncGenerator, AsyncIterator, List, Optional, Union
 
 import litellm
 
-from llama_stack import logcat
 from llama_stack.apis.common.content_types import (
     InterleavedContent,
     InterleavedContentItem,
@@ -33,6 +32,7 @@ from llama_stack.apis.inference import (
 )
 from llama_stack.apis.models.models import Model
 from llama_stack.distribution.request_headers import NeedsRequestProviderData
+from llama_stack.log import get_logger
 from llama_stack.providers.utils.inference.model_registry import (
     ModelRegistryHelper,
 )
@@ -47,6 +47,8 @@ from llama_stack.providers.utils.inference.prompt_adapter import (
     interleaved_content_as_str,
 )
 
+logger = get_logger(name=__name__, category="inference")
+
 
 class LiteLLMOpenAIMixin(
     ModelRegistryHelper,
@@ -74,7 +76,7 @@ class LiteLLMOpenAIMixin(
         self,
         model_id: str,
         content: InterleavedContent,
-        sampling_params: Optional[SamplingParams] = SamplingParams(),
+        sampling_params: Optional[SamplingParams] = None,
         response_format: Optional[ResponseFormat] = None,
         stream: Optional[bool] = False,
         logprobs: Optional[LogProbConfig] = None,
@@ -85,7 +87,7 @@ class LiteLLMOpenAIMixin(
         self,
         model_id: str,
         messages: List[Message],
-        sampling_params: Optional[SamplingParams] = SamplingParams(),
+        sampling_params: Optional[SamplingParams] = None,
         tools: Optional[List[ToolDefinition]] = None,
         tool_choice: Optional[ToolChoice] = ToolChoice.auto,
         tool_prompt_format: Optional[ToolPromptFormat] = None,
@@ -94,6 +96,8 @@ class LiteLLMOpenAIMixin(
         logprobs: Optional[LogProbConfig] = None,
         tool_config: Optional[ToolConfig] = None,
     ) -> Union[ChatCompletionResponse, AsyncIterator[ChatCompletionResponseStreamChunk]]:
+        if sampling_params is None:
+            sampling_params = SamplingParams()
         model = await self.model_store.get_model(model_id)
         request = ChatCompletionRequest(
             model=model.provider_resource_id,
@@ -107,8 +111,7 @@ class LiteLLMOpenAIMixin(
         )
 
         params = await self._get_params(request)
-        logcat.debug("inference", f"params to litellm (openai compat): {params}")
-
+        logger.debug(f"params to litellm (openai compat): {params}")
         # unfortunately, we need to use synchronous litellm.completion here because litellm
         # caches various httpx.client objects in a non-eventloop aware manner
         response = litellm.completion(**params)

llama_stack/providers/utils/inference/openai_compat.py

@@ -615,6 +615,14 @@ def convert_tool_call(
     return valid_tool_call
 
 
+PYTHON_TYPE_TO_LITELLM_TYPE = {
+    "int": "integer",
+    "float": "number",
+    "bool": "boolean",
+    "str": "string",
+}
+
+
 def convert_tooldef_to_openai_tool(tool: ToolDefinition) -> dict:
     """
     Convert a ToolDefinition to an OpenAI API-compatible dictionary.
@@ -675,7 +683,7 @@ def convert_tooldef_to_openai_tool(tool: ToolDefinition) -> dict:
         properties = parameters["properties"]
         required = []
         for param_name, param in tool.parameters.items():
-            properties[param_name] = {"type": param.param_type}
+            properties[param_name] = {"type": PYTHON_TYPE_TO_LITELLM_TYPE.get(param.param_type, param.param_type)}
             if param.description:
                 properties[param_name].update(description=param.description)
             if param.default:

llama_stack/providers/utils/inference/prompt_adapter.py

@@ -8,14 +8,12 @@ import asyncio
 import base64
 import io
 import json
-import logging
 import re
 from typing import List, Optional, Tuple, Union
 
 import httpx
 from PIL import Image as PIL_Image
 
-from llama_stack import logcat
 from llama_stack.apis.common.content_types import (
     ImageContentItem,
     InterleavedContent,
@@ -34,6 +32,7 @@ from llama_stack.apis.inference import (
     ToolDefinition,
     UserMessage,
 )
+from llama_stack.log import get_logger
 from llama_stack.models.llama.datatypes import (
     ModelFamily,
     RawContent,
@@ -58,7 +57,7 @@ from llama_stack.models.llama.llama3.tokenizer import Tokenizer
 from llama_stack.models.llama.sku_list import resolve_model
 from llama_stack.providers.utils.inference import supported_inference_models
 
-log = logging.getLogger(__name__)
+log = get_logger(name=__name__, category="inference")
 
 
 class ChatCompletionRequestWithRawContent(ChatCompletionRequest):
@@ -464,7 +463,7 @@ def _get_tool_choice_prompt(tool_choice: ToolChoice | str, tools: List[ToolDefin
 def get_default_tool_prompt_format(model: str) -> ToolPromptFormat:
     llama_model = resolve_model(model)
     if llama_model is None:
-        logcat.warning("inference", f"Could not resolve model {model}, defaulting to json tool prompt format")
+        log.warning(f"Could not resolve model {model}, defaulting to json tool prompt format")
         return ToolPromptFormat.json
 
     if llama_model.model_family == ModelFamily.llama3_1 or (