move is_hosted out of the NVIDIAConfig api

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

@@ -46,7 +46,3 @@ class NVIDIAConfig(BaseModel):
         default=60,
         description="Timeout for the HTTP requests",
     )
-
-    @property
-    def is_hosted(self) -> bool:
-        return "integrate.api.nvidia.com" in self.url

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

@@ -40,7 +40,7 @@ from ._openai_utils import (
     convert_openai_chat_completion_choice,
     convert_openai_chat_completion_stream,
 )
-from ._utils import check_health
+from ._utils import _is_nvidia_hosted, check_health
 
 _MODEL_ALIASES = [
     build_model_alias_with_just_provider_model_id(
@@ -91,7 +91,7 @@ class NVIDIAInferenceAdapter(Inference, ModelRegistryHelper):
 
         print(f"Initializing NVIDIAInferenceAdapter({config.url})...")
 
-        if config.is_hosted:
+        if _is_nvidia_hosted(config):
             if not config.api_key:
                 raise RuntimeError(
                     "API key is required for hosted NVIDIA NIM. "

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

@@ -11,6 +11,10 @@ import httpx
 from ._config import NVIDIAConfig
 
 
+def _is_nvidia_hosted(config: NVIDIAConfig) -> bool:
+    return "integrate.api.nvidia.com" in config.url
+
+
 async def _get_health(url: str) -> Tuple[bool, bool]:
     """
     Query {url}/v1/health/{live,ready} to check if the server is running and ready
@@ -37,7 +41,7 @@ async def check_health(config: NVIDIAConfig) -> None:
     Raises:
         RuntimeError: If the server is not running or ready
     """
-    if not config.is_hosted:
+    if not _is_nvidia_hosted(config):
         print("Checking NVIDIA NIM health...")
         try:
             is_live, is_ready = await _get_health(config.url)