phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-08-03 17:29:01 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

move is_hosted out of the NVIDIAConfig api

Browse source
This commit is contained in:
Matthew Farrellee 2024-11-21 15:08:31 -05:00
parent 988741c276
commit 8944491c3c
3 changed files with 7 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

4
llama_stack/providers/remote/inference/nvidia/_config.py
View file

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

4
llama_stack/providers/remote/inference/nvidia/_nvidia.py
View file

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

6
llama_stack/providers/remote/inference/nvidia/_utils.py
View file

@ -11,6 +11,10 @@ import httpx
from ._config import NVIDIAConfig 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]: async def _get_health(url: str) -> Tuple[bool, bool]:
""" """
Query {url}/v1/health/{live,ready} to check if the server is running and ready 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: Raises:
RuntimeError: If the server is not running or ready 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...") print("Checking NVIDIA NIM health...")
try: try:
is_live, is_ready = await _get_health(config.url) is_live, is_ready = await _get_health(config.url)