Chad Lee

I'm not sure why it would not be including the built files in the optimizer build. I don't really use this plugin anymore and don't have much time to look...

> GPT-4 might be less brittle to this unorthodox use of user messages, but the performance of open source models degrade a lot when the expected format is not used....

my main point was the "is this true?" - the positioning of the messages leading to "suboptimal performance" and if so, how suboptimal?

I don't really use this library anymore and don't have the energy to put development time into it. I'm happy to merge a PR to fix issues like this though,...

definitely not going to manage multiple versions. but I think doing a major version upgrade with support for newer bootstrap versions makes sense going forward. in other words, forget about...

heh, yeah I wrote that line 2 years ago :sweat_smile: haven't kept up-to-date with the latest to know if that line even makes sense anymore

Hmm, wouldn't this load & execute the worker script in the context of the calling page?