We already know ChatGPT is not good enough to do all that without being able to run code. That's why ChatGPT Code Interpreter is so powerful. It was even branded GPT4.5 in this podcast[0] which I think is apt.
However, bringing that auto-coding loop into a persistent executable environment that you can revisit, adjust, and follow up on is a big win I think.
Hey HN! That's me talking in the video. This is basically GPT-4 + the new functions capability. We added this to Deepnote during a 2-day hackathon and were surprised by how well it works. Let me know if you have any questions!
Thanks for the feedback! The main difference between us (Deepnote) and Colab is that we aim for a clean reading experience, and we optimize for speed.
Colab seems to load a notebook that has editable cells and you can start executing cells directly, so it doesn't feel as a "publishing" feature, more like straight up notebook sharing.
And thanks for the bug report, will look into that
Hi there! Simon (Engineer at Deepnote) here. We didn't benchmark the load speed but for some reason, Github's ipynb viewer has always felt to me quite slow and unreliable.
All viewers are publicly accessible so I'd love it someone did an independent benchmark. I'd prefer to avoid doing one ourselves because of the obvious conflict of interest
Deepnote team member here. It took a lot of effort to get to where we are right now – Deepnote is one of the most complex products I worked on. If you have any questions, engineering, product or otherwise, ask away!