Yes I have base css/js that I inject on top of whatever codegen gemini 3 comes back with -- It runs via ai-sdk so the specific function is streamObject which is prompted to generate inner HTML elements

Thanks! Half of the work here was getting gemini to generate reasonable canvas animations within the window. I'm still experimenting which "style" I should keep/add

Huge- good to hear will see if I can add a way to type in what you want or to go dig further into just certain hashtags.

One way to get usable and consistent categories for a Wikipedia article is Lift Wing: https://wikitech.wikimedia.org/wiki/Machine_Learning/LiftWin...

I once used it to create a "Wikipedia StumbleUpon": https://github.com/fpsvogel/wiki-stumble#demo

Here is the code related to Lift Wing: https://github.com/fpsvogel/wiki-stumble/blob/main/app/model...

Mentioning this in case you're looking for something as a basis for tags.

EDIT: Here is a list of all the categories from Lift Wing: https://github.com/fpsvogel/wiki-stumble/blob/main/app/model...

It's pretty cool.

This is even scalable if you consider caching some of these requests and allow users to choose if they want a just-in-time version. Also the comments really are snappy and quick to load.

Yes just fixed timer, and using browser TTS nothing fancy here on purpose - when I did some research on tiktok videos generally simpler/worse quality seemed to be better XD

In this case, static text without the highlighter would have been better. Because the timing is so far off, the highlighter interferes with reading the captions for hearing assistance.

To prevent overfitting

Getting there! training a new model, on thinking machines' tinker api which uses a considerably larger and different architecture.

In the meanwhile I tried at least 100+ variations of trying to train this model to be SOTA most of which led me down getting better data which I believe I do, just needs more tuning still.

this is great, outcome based pricing is definitely a competitive advantage and more aligns you with your customer!

100%! We think pricing and biz model will become a differentiator. Being able to adapt and iterate on pricing will be really important as underlying models get better

thats great -- what TPS were you getting? github: https://github.com/sdan/r1-web

the filter between slop and good music increases