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

add comment todos

Browse source
This commit is contained in:
Xi Yan 2024-09-09 11:50:35 -07:00
parent 8c378fadcc
commit 26209a9d99
2 changed files with 2 additions and 14 deletions
Download patch file Download diff file Expand all files Collapse all files

13
llama_toolchain/agentic_system/client_sdk.py
View file

@ -7,17 +7,6 @@ def main(host: str, port: int):
base_url=f"http://{host}:{port}", base_url=f"http://{host}:{port}",
) )
# Need smt like this to work w/ server, however this is not what was generated by SDK
# 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": True,
# },
# )
agentic_system_create_response = client.agentic_system.create( agentic_system_create_response = client.agentic_system.create(
agent_config={ agent_config={
"instructions": "You are a helpful assistant", "instructions": "You are a helpful assistant",
@ -32,7 +21,7 @@ def main(host: str, port: int):
) )
print(agentic_system_create_session_response) print(agentic_system_create_session_response)
# TODO(xiyan): This does not work with current server, need to wrap it in a request (similar to AgentConfig?) # TODO(xiyan): remove request wrapper
response = client.agentic_system.turns.create( response = client.agentic_system.turns.create(
request={ request={
"agent_id": agentic_system_create_response.agent_id, "agent_id": agentic_system_create_response.agent_id,

3
llama_toolchain/inference/client_sdk.py
View file

@ -7,7 +7,6 @@ def main(host: str, port: int):
base_url=f"http://{host}:{port}", base_url=f"http://{host}:{port}",
) )
# Need smt like this to work w/ server, however this is not what was generated by SDK (?)
response = client.inference.chat_completion( response = client.inference.chat_completion(
request={ request={
"messages": [ "messages": [
@ -19,7 +18,7 @@ def main(host: str, port: int):
) )
print(response) print(response)
# This does not work with current server # TODO (xiyan). This does not work with current server, need to fix
# response = client.inference.chat_completion( # response = client.inference.chat_completion(
# messages=[ # messages=[
# UserMessage(content="hello world, troll me in two-paragraphs about 42", role="user"), # UserMessage(content="hello world, troll me in two-paragraphs about 42", role="user"),