AutoGen Studio needs to support main/AutoGen 0.4

[dsanr]

What happened?

Support for human input is merged with the main branch #3484
When autogenstudio is installed with 'pip install autogenstudio', it is not accessing the main branch, even if the main branch is installed locally.
Autogenstudio is accessing only a different package, which is getting installed from pip, that doesn't include the mentioned PR.

How can I use Human Input Mode in AutoGen Studio? Do we need a separate release?

What did you expect to happen?

Human Input Mode in AutoGen Studio should work when installing the main branch or 0.4 dev1 branch

How can we reproduce it (as minimally and precisely as possible)?

pip install autogen-agentchat==0.4.0dev1
pip install autogenstudio
autogenstudio ui

Then try the options of Human_Input_Mode

AutoGen version

0.4

Which package was this bug in

AutoGen Studio

Model used

No response

Python version

No response

Operating system

No response

Any additional info you think would be helpful for fixing this bug

No response

[rysweet]

@dsanr - Autogen studio does not yet support AutoGen 0.4. We do plan to revise it to support 0.4 but do not have an ETA. Contributions welcome.

[dsanr]

@rysweet Thank you for editing the issue and reply.
By the way, how is https://github.com/autogen-ai/autogen different from this repository? Which one is likely to get frequent updates?

[rysweet]

@rysweet Thank you for editing the issue and reply. By the way, how is https://github.com/autogen-ai/autogen different from this repository? Which one is likely to get frequent updates?

@dsanr - please see the FAQs: https://github.com/microsoft/autogen?tab=readme-ov-file#faqs
In short - the other project is a fork that isn't affiliated with the official autogen or microsoft. that fork also uses the old autogen arch that we have moved on from for the reasons cited in the blog post about the new arch. We believe that the new event-driven arch is more suitable for production workloads at scale and offers a more flexible programming model.