I build ChatGPT-2D with the idea to enable people to branch AI conversations interactively, pose contextual questions based on AI response and visualize your entire dialogue on a 2 dimensional map interface. What do you think?
It looks similar to loom <a href="https://github.com/socketteer/loom">https://github.com/socketteer/loom</a> the program that was used in the research behind the simulators essay <a href="https://generative.ink/posts/simulators/" rel="nofollow">https://generative.ink/posts/simulators/</a>
This is a really good idea, I wish I had thought of it. Thinking of chat as a graph instead of a list is very weird and very cool.<p>I think this could be especially useful in a research context where someone is trying to evaluate the performance of a model. Sometimes you just want to compare how a model responds with and without certain context, and doing so with the classic UI is cumbersome.
You have probably one of the best landing pages I have seen in a long time for a new product. There's no BS, in 10 seconds I can understand what the idea is about. Well done!
I've been working on something very similar in the past three weeks. Just out of curiosity, what have you used for the layout?<p>In my project I used reactflow for all things graph and dagre for the layouting (together with some custom hooks to make it work with variable size nodes).<p>I'm not sure if I will try to monetise it or if I will simply open-source it. Probably the latter because when building time is over and marketing time starts, I get very bored and lose interest soon.
I'm really enjoying the Superus-101 playground for newbies! I've been playing around with it for a while now, and I find it really helpful for brainstorming. However, I did notice that the waiting time for a response can be a bit long. Do you think it would be possible to implement streaming and concurrent modes for generating new nodes in the graph?
Very cool! Could definitely see myself using this. Out of curiosity, what's the visualization library you're using to generate the boxes/lines/zoom in-out canvas? Or is that all native to your app? Building something unrelated and looking for something like this
Man this is cool.<p>I did this manually myself to explore some topics I was curious about: I'd copy&paste responses into a Google doc, and then add sub and subsubsections with answers to sub-questions.<p>Will give it a spin!
Whimsical, a diagram tool, has a new feature that works a little like this but you have to do a bit more work, I've been enjoying it. I'll have to give this a go, thanks op!
I like this. I've done something similar by asking chatgpt to map out some ideas for me. It knows how to output into the `dot` format which makes the result easy to post-process.