Can anyone share the vision models they might be using and how they might be tuning them?

GitHub teased a new functionality called Copilot workspace in GitHub universe 2023.

My guess is this is going to be powered using GPT-5/4.5 and they might've even used the model in the demo.

It's bs. OP has only copy pasted what we all can read. Not the subscriber part.

The thing with langchain is that it's documentation is terrible.

If I want to do something I have to rely on other sources to use it. That's just an overkill.

Use majorly for programming and also to learn something by simplifying or digging deeper using questions.

Chatgpt as a debugging partner has been a adam and eve sort of match to me. It's a saviour!

Is seems disruptive but for current users there isn't something new here.

Also GPT-4 is just optimisations around speed. That's a big overkill of everyone's expectations.