Scrapped the llama2.ts file in favor of openrouter Added openrouter support Worked on the basis for model swapping Rethought a bit of code I had in place Switched GPT4 back to turbo so I can use functions Began integration of VDB (GPT4 only for now while I get my shit together)
Added GPT-4 Vision for those who like it did formatting uhh the uhh fixed VDB did some magijcks to the rest of the files idk man I forgor
Reworked how LLMs are loaded, the bot is less of a full solution and more of a framework now