Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Feature Request]: Structured Output/ Formatted Output #25

Open
Hk669 opened this issue Sep 3, 2024 · 6 comments
Open

[Feature Request]: Structured Output/ Formatted Output #25

Hk669 opened this issue Sep 3, 2024 · 6 comments
Labels
enhancement New feature or request roadmap Work planned

Comments

@Hk669
Copy link
Collaborator

Hk669 commented Sep 3, 2024

Is your feature request related to a problem? Please describe.

Ive came across many comments from the community, asking for the formatted output/ structural output from the Agentic chat (initate_chat and groupchat ). I believe generating the expected response after the agentic chat can be the biggest improvement in AutoGen.

Me, @marklysze and @sonichi discussed about adding this feature over a call.

Describe the solution you'd like

There should be an output message after the completion of the chat, i can see there is last_message, but that doesn't serve the exact feature. adding this feature can help production apps use AutoGen.

solution: TBD ( help and ideas are appreciated )

Additional context

We can get started with the openai structured output and final goal should be the structured output for the complete chat.

cc @sonichi @BeibinLi @qingyun-wu @marklysze @yiranwu0 @skzhang1

@Hk669 Hk669 added enhancement New feature or request roadmap Work planned labels Sep 3, 2024
@davorrunje
Copy link
Collaborator

I already implemented one version of structured input/output messages for internal usage. I could take this one.

@r4881t
Copy link
Collaborator

r4881t commented Sep 6, 2024

Another feature request and discussion on the same lines. @davorrunje @Hk669 microsoft#3442 @davorrunje i would love to help, please let me know if i can. My discord is xpranav on Autogen server

@marklysze
Copy link
Collaborator

Great to get a discussion here on approaches (and @davorrunje what you are doing already) and see if we can come to a consensus on a way forward.

@Hk669
Copy link
Collaborator Author

Hk669 commented Sep 7, 2024

Great to get a discussion here on approaches (and @davorrunje what you are doing already) and see if we can come to a consensus on a way forward.

it's at fastagency.ai, I believe it's only using the chat.summary() method for the output. correct me if I'm wrong. @davorrunje

@marklysze
Copy link
Collaborator

Feel free to put down what you'd like to see and how you could see it working. OpenAI's Structured Outputs is a starter.

@tylerprogramming
Copy link
Collaborator

@davorrunje Hey! Just curious if there was any update to this or if it's still being worked on?

odoochain pushed a commit to odoochain/autogen that referenced this issue Nov 10, 2024
* make ghClient fetch synchronous

* refactor memory, extract knowledge adding to AiAgent
odoochain pushed a commit to odoochain/autogen that referenced this issue Nov 10, 2024
…responses (autogenhub#25)

* rename broadcast to publish

* remove require response, remove responses from publishing
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request roadmap Work planned
Projects
None yet
Development

No branches or pull requests

5 participants