mirror of
https://github.com/meta-llama/llama-stack.git
synced 2025-12-17 07:02:36 +00:00
34 lines
951 B
Python
34 lines
951 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.
|
|
|
|
import os
|
|
|
|
from llama_stack_client import LlamaStackClient
|
|
|
|
from pydantic import BaseModel
|
|
|
|
client = LlamaStackClient(base_url=f"http://localhost:{os.environ['LLAMA_STACK_PORT']}")
|
|
|
|
|
|
class CompletionMessage(BaseModel):
|
|
recipe_name: str
|
|
ingredients: list[str]
|
|
steps: list[str]
|
|
|
|
|
|
response = client.inference.chat_completion(
|
|
model_id=os.environ["INFERENCE_MODEL"],
|
|
messages=[
|
|
{"role": "system", "content": "You are a chef."},
|
|
{"role": "user", "content": "Give me a recipe for spaghetti bolognaise"},
|
|
],
|
|
response_format={
|
|
"type": "json_schema",
|
|
"json_schema": CompletionMessage.model_json_schema(),
|
|
},
|
|
sampling_params={"max_tokens": 8000},
|
|
)
|
|
print(response.completion_message.content)
|