forked from phoenix-oss/llama-stack-mirror
4d0bfbf984
This PR adds a llama-stack inference provider for `api.llama.com`, as well as adds entries for Llama-Guard-4 and updated Prompt-Guard models.
34 lines
1,018 B
Python
34 lines
1,018 B
Python
# Copyright (c) Meta Platforms, Inc. and affiliates.
|
|
# All rights reserved.
|
|
#
|
|
# This source code is licensed under the terms described in the LICENSE file in
|
|
# the root directory of this source tree.
|
|
|
|
from llama_stack.providers.remote.inference.llama_openai_compat.config import (
|
|
LlamaCompatConfig,
|
|
)
|
|
from llama_stack.providers.utils.inference.litellm_openai_mixin import (
|
|
LiteLLMOpenAIMixin,
|
|
)
|
|
|
|
from .models import MODEL_ENTRIES
|
|
|
|
|
|
class LlamaCompatInferenceAdapter(LiteLLMOpenAIMixin):
|
|
_config: LlamaCompatConfig
|
|
|
|
def __init__(self, config: LlamaCompatConfig):
|
|
LiteLLMOpenAIMixin.__init__(
|
|
self,
|
|
model_entries=MODEL_ENTRIES,
|
|
api_key_from_config=config.api_key,
|
|
provider_data_api_key_field="llama_api_key",
|
|
openai_compat_api_base=config.openai_compat_api_base,
|
|
)
|
|
self.config = config
|
|
|
|
async def initialize(self):
|
|
await super().initialize()
|
|
|
|
async def shutdown(self):
|
|
await super().shutdown()
|