phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-07-29 15:23:51 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

remove client sdk examples

Browse source
This commit is contained in:
Xi Yan 2024-09-09 12:21:08 -07:00
parent 26209a9d99
commit 6bfcbc678e
2 changed files with 0 additions and 71 deletions
Download patch file Download diff file Expand all files Collapse all files

40
llama_toolchain/agentic_system/client_sdk.py
View file

@ -1,40 +0,0 @@
import fire
from llama_stack import LlamaStack
from llama_stack.types import UserMessage
def main(host: str, port: int):
client = LlamaStack(
base_url=f"http://{host}:{port}",
)
agentic_system_create_response = client.agentic_system.create(
agent_config={
"instructions": "You are a helpful assistant",
"model": "Meta-Llama3.1-8B-Instruct",
},
)
print(agentic_system_create_response)
agentic_system_create_session_response = client.agentic_system.sessions.create(
agent_id=agentic_system_create_response.agent_id,
session_name="test_session",
)
print(agentic_system_create_session_response)
# TODO(xiyan): remove request wrapper
response = client.agentic_system.turns.create(
request={
"agent_id": agentic_system_create_response.agent_id,
"session_id": agentic_system_create_session_response.session_id,
"messages": [
UserMessage(content="What is the capital of France?", role="user"),
],
"stream": False,
}
)
print(response)
if __name__ == "__main__":
fire.Fire(main)

31
llama_toolchain/inference/client_sdk.py
View file

@ -1,31 +0,0 @@
import fire
from llama_stack import LlamaStack
from llama_stack.types import UserMessage
def main(host: str, port: int):
client = LlamaStack(
base_url=f"http://{host}:{port}",
)
response = client.inference.chat_completion(
request={
"messages": [
UserMessage(content="hello world, troll me in two-paragraphs about 42", role="user"),
],
"model": "Meta-Llama3.1-8B-Instruct",
"stream": False,
},
)
print(response)
# TODO (xiyan). This does not work with current server, need to fix
# response = client.inference.chat_completion(
# messages=[
# UserMessage(content="hello world, troll me in two-paragraphs about 42", role="user"),
# ],
# model="Meta-Llama3.1-8B-Instruct",
# stream=True,
# )
if __name__ == "__main__":
fire.Fire(main)