phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-10-08 13:00:52 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

chore: Revert "fix: fix nvidia provider (#3716)" (#3730)

Browse source 
This reverts commit c940fe7938.

@wukaixingxp I stamped to fast. Let's wait for @mattf's review.
This commit is contained in:
ehhuang 2025-10-07 19:16:51 -07:00 • committed by GitHub
parent c940fe7938
commit b6e9f41041
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 7 additions and 21 deletions
Download patch file Download diff file Expand all files Collapse all files

28
llama_stack/providers/remote/inference/nvidia/nvidia.py
View file

@ -13,7 +13,6 @@ from llama_stack.apis.inference import (
OpenAIEmbeddingUsage,
)
from llama_stack.log import get_logger
from llama_stack.providers.utils.inference.model_registry import ModelRegistryHelper
from llama_stack.providers.utils.inference.openai_mixin import OpenAIMixin
from . import NVIDIAConfig
@ -22,7 +21,9 @@ from .utils import _is_nvidia_hosted
logger = get_logger(name=__name__, category="inference::nvidia")
class NVIDIAInferenceAdapter(OpenAIMixin, ModelRegistryHelper):
class NVIDIAInferenceAdapter(OpenAIMixin):
config: NVIDIAConfig
"""
NVIDIA Inference Adapter for Llama Stack.
@ -36,27 +37,12 @@ class NVIDIAInferenceAdapter(OpenAIMixin, ModelRegistryHelper):
- ModelRegistryHelper.check_model_availability() just returns False and shows a warning
"""
def __init__(self, config: NVIDIAConfig) -> None:
"""Initialize the NVIDIA inference adapter with configuration."""
# Initialize ModelRegistryHelper with empty model entries since NVIDIA uses dynamic model discovery
ModelRegistryHelper.__init__(self, model_entries=[], allowed_models=config.allowed_models)
self.config = config
# source: https://docs.nvidia.com/nim/nemo-retriever/text-embedding/latest/support-matrix.html
embedding_model_metadata: dict[str, dict[str, int]] = {
"nvidia/llama-3.2-nv-embedqa-1b-v2": {
"embedding_dimension": 2048,
"context_length": 8192,
},
"nvidia/llama-3.2-nv-embedqa-1b-v2": {"embedding_dimension": 2048, "context_length": 8192},
"nvidia/nv-embedqa-e5-v5": {"embedding_dimension": 512, "context_length": 1024},
"nvidia/nv-embedqa-mistral-7b-v2": {
"embedding_dimension": 512,
"context_length": 4096,
},
"snowflake/arctic-embed-l": {
"embedding_dimension": 512,
"context_length": 1024,
},
"nvidia/nv-embedqa-mistral-7b-v2": {"embedding_dimension": 512, "context_length": 4096},
"snowflake/arctic-embed-l": {"embedding_dimension": 512, "context_length": 1024},
}
async def initialize(self) -> None:
@ -109,7 +95,7 @@ class NVIDIAInferenceAdapter(OpenAIMixin, ModelRegistryHelper):
response = await self.client.embeddings.create(
model=await self._get_provider_model_id(model),
input=input,
encoding_format=(encoding_format if encoding_format is not None else NOT_GIVEN),
encoding_format=encoding_format if encoding_format is not None else NOT_GIVEN,
dimensions=dimensions if dimensions is not None else NOT_GIVEN,
user=user if user is not None else NOT_GIVEN,
extra_body=extra_body,