gents, does the OpenAIAgent (the one

RRoland Tannous

gents, does the OpenAIAgent (the one referenced here https://docs.llamaindex.ai/en/stable/examples/agent/openai_agent/) keep conversation history? it doesn't seem to do so. What's the best way to go about it?

43 comments

LLogan M

it sure does

LLogan M

agent.memory.get() will get the current buffer

LLogan M

Plain Text

>>> from llama_index.agent.openai import OpenAIAgent
>>> agent = OpenAIAgent.from_tools([])
>>> resp = agent.chat("My name is logan")
>>> print(str(resp))
Nice to meet you, Logan! How can I assist you today?
>>> resp = agent.chat("I forgot my name, what was it?")
>>> print(str(resp))
Your name is Logan. Is there anything else I can help you with?
>>>

RRoland Tannous

hey Logan. Tahnk you for the answer.
It seems the issue is not caused by OpenAIAgent itself (I initially thought it was).
The real issue has to do with streamlit's str.chatinput function. It's somewhat stateless ? and seems to cause conversation memory to get lost . so i'm trying to work around that now.
also this related thread:
https://discord.com/channels/1059199217496772688/1240358158116061195

LLogan M

You probably need to put the agent into the session_state of the app.

Either that, or manually keep track of and insert the chat history on each .chat() call to override anything in the memory

Plain Text

chat_history = agent.chat_history
agent.chat("Hello!", chat_history=chat_history)

RRoland Tannous

oh 🤔

RRoland Tannous

let me try it out.

RRoland Tannous

didn't work but i had another turnaround work here. let me just clean the code and share the snippet of what worked.

RRoland Tannous

again more a streamlit issue than llamaindex

RRoland Tannous

ok so what works for me is this

RRoland Tannous

this

RRoland Tannous

Plain Text

@st.cache_resource
def initialize_agent():
    agent = get_agent()
    return agent

agent = initialize_agent()

RRoland Tannous

in the same file where i am using the streamlit chat widget

RRoland Tannous

get_agent simply grabs the OpenAIAPI agent that is being defined in a separate python help file.

RRoland Tannous

originally while looking for a solution tracked this thread on langchain github

RRoland Tannous

https://github.com/langchain-ai/langchain/issues/7061

RRoland Tannous

but i didn't want to use their component, .... StreamlitChatMessageHistory

RRoland Tannous

then i thought that maybe finding a way to initialize the agent only once, and not have it reset everytime the page runs could solve it, so i found this thread: https://discuss.streamlit.io/t/is-there-a-way-to-run-an-initialization-function/61154/3

RRoland Tannous

which took me here:

RRoland Tannous

https://docs.streamlit.io/develop/api-reference/caching-and-state/st.cache_resource

RRoland Tannous

using the @st.cache_resource decorator seems to solve my problem

RRoland Tannous

took me a whole day 😐

RRoland Tannous

thank god we don't use streamlit in production 🤣

RRoland Tannous

quick question though @Logan M accordint to this page, the OpenAIAgent should be able to do streaming...

https://docs.llamaindex.ai/en/stable/examples/agent/openai_agent/#agent-with-personality

LLogan M

yea, agent.stream_chat()

RRoland Tannous

however whenever i do agent.stream_chat() it's throwing an error

LLogan M

Whats the error?

RRoland Tannous

streaming works with other Agent Types like ReacTagent, and the agentrunner / agentworker combo fine.... only errors on OpenAIAgent

RRoland Tannous

Plain Text

AttributeError: 'StreamingAgentChatResponse' object has no attribute '_is_function_not_none_thread_event'

RRoland Tannous

😐

RRoland Tannous

on this line

RRoland Tannous

Plain Text

response = agent.stream_chat(prompt)

RRoland Tannous

llama-index 0.10.37
llama-index-agent-openai 0.2.2

RRoland Tannous

llama-index-llms-openai 0.1.16

RRoland Tannous

This seems to be mentioned here by jeremielenoir:
https://github.com/run-llama/llama_index/issues/11708

RRoland Tannous

but can't figure out the connection 😄

LLogan M

Hmm, this should be just a versioning thing. Try pip install -U llama-index-agent-openai llama-index-core llama-index-llms-openai just to be sure? (You might have to restart anything you have running too)

RRoland Tannous

Plain Text

agent = OpenAIAgent.from_tools([vector_query_tool, *googletools],llm=llm,verbose=True,system_prompt=get_system_prompt())

That's how agent is defined

RRoland Tannous

oh let me try that

RRoland Tannous

moment of truth 🥁

RRoland Tannous

yes it was a versioning issue.
Works perfectly with upgraded versions:
llama-index-agent-openai-0.2.5 llama-index-llms-openai-0.1.19

RRoland Tannous

😏

cconceptt

thanks from the future!

Add a reply

Find answers from the community

gents, does the OpenAIAgent (the one